Viewing docs for konnect 3.23.0
published on Friday, Sep 18, 2026 by kong
published on Friday, Sep 18, 2026 by kong
Viewing docs for konnect 3.23.0
published on Friday, Sep 18, 2026 by kong
published on Friday, Sep 18, 2026 by kong
AIGatewayModel Resource
Example Usage
import * as pulumi from "@pulumi/pulumi";
import * as konnect from "@pulumi/konnect";
const myAigatewaymodel = new konnect.AiGatewayModel("my_aigatewaymodel", {
api: {
access: {
acls: {
allows: ["..."],
denies: ["..."],
},
authStrategies: ["okta-ai-se"],
identityProviders: ["okta-ai-se"],
},
capabilities: ["files"],
config: {
balancer: {
lowestLatency: {
connectTimeout: 60000,
failTimeout: 10000,
failoverCriterias: ["http_429"],
latencyStrategy: "tpot",
maxFails: 0,
readTimeout: 60000,
retries: 5,
slots: 10000,
writeTimeout: 60000,
},
},
logging: {
payloads: false,
},
maxRequestBodySize: 8388608,
proxy: {
auth: {
password: "...my_password...",
username: "...my_username...",
},
httpProxy: {
host: "...my_host...",
port: 29747,
},
httpsProxy: {
host: "...my_host...",
port: 12764,
},
noProxy: "...my_no_proxy...",
proxyScheme: "http",
},
responseStreaming: "allow",
route: {
headers: {
key: JSON.stringify("value"),
},
hosts: ["foo.example.com"],
httpsRedirectStatusCode: 426,
methods: ["..."],
model: {
bodyParam: "model",
headerParam: "x-model",
pathParam: "model_name",
values: ["..."],
},
paths: ["..."],
preserveHost: false,
protocols: ["..."],
regexPriority: 0,
requestBuffering: true,
responseBuffering: true,
stripPath: true,
tags: ["..."],
},
},
displayName: "My GPT 5 model",
enabled: true,
formats: [{
type: "openai",
}],
labels: {
key: "value",
},
managedBy: {
key: "value",
},
name: "my-gpt-5-model",
policies: ["..."],
targets: [{
allowAuthOverride: false,
config: {
xai: {
cacheReadCost: 4.42,
cacheWriteCost: 3.9,
cacheWriteCostLists: [{
cost: 7.32,
ttl: "...my_ttl...",
}],
contextWindowFactors: [{
above: "...my_above...",
inputFactor: 1.42,
outputFactor: 1.31,
}],
embeddingsDimensions: 1556463673,
inputCost: 3.7,
maxTokens: 1227329724,
outputCost: 6.56,
serviceTierFactors: [{
factor: 8.57,
tier: "...my_tier...",
}],
temperature: 3.27,
topK: 483136424,
topP: 2.83,
upstreamUrl: "https://baggy-trash.biz/",
},
},
name: "gpt-5-model",
provider: "azure-ai-se",
semanticDescription: "...my_semantic_description...",
weight: 100,
}],
},
gatewayId: "5f9fd312-a987-4628-b4c5-bb4f4fddd5f7",
model: {
access: {
acls: {
allows: ["..."],
denies: ["..."],
},
authStrategies: ["okta-ai-se"],
identityProviders: ["okta-ai-se"],
},
capabilities: ["audio/speech"],
config: {
balancer: {
semantic: {
connectTimeout: 60000,
embeddings: {
allowAuthOverride: false,
config: {
ollama: {
upstreamUrl: "...my_upstream_url...",
},
},
name: "...my_name...",
provider: "azure-ai-se",
},
failTimeout: 10000,
failoverCriterias: ["http_502"],
maxFails: 0,
readTimeout: 60000,
retries: 5,
slots: 10000,
vectordb: {
pgvector: {
database: "kong-pgvector",
dimensions: 6,
distanceMetric: "euclidean",
host: "127.0.0.1",
password: "...my_password...",
port: 5432,
ssl: {
cert: "...my_cert...",
certKey: "...my_cert_key...",
enabled: true,
required: true,
verify: true,
version: "tlsv1_2",
},
threshold: 3.66,
timeout: 5000,
user: "postgres",
},
},
writeTimeout: 60000,
},
},
logging: {
payloads: false,
},
maxRequestBodySize: 8388608,
model: {
nameHeader: true,
},
proxy: {
auth: {
password: "...my_password...",
username: "...my_username...",
},
httpProxy: {
host: "...my_host...",
port: 30633,
},
httpsProxy: {
host: "...my_host...",
port: 29606,
},
noProxy: "...my_no_proxy...",
proxyScheme: "http",
},
responseStreaming: "allow",
route: {
headers: {
key: JSON.stringify("value"),
},
hosts: ["foo.example.com"],
httpsRedirectStatusCode: 426,
methods: ["..."],
model: {
bodyParam: "model",
headerParam: "x-model",
pathParam: "model_name",
values: ["..."],
},
paths: ["..."],
preserveHost: false,
protocols: ["..."],
regexPriority: 0,
requestBuffering: true,
responseBuffering: true,
stripPath: true,
tags: ["..."],
},
},
displayName: "My GPT 5 model",
enabled: true,
formats: [{
type: "openai",
}],
labels: {
key: "value",
},
managedBy: {
key: "value",
},
name: "my-gpt-5-model",
policies: ["..."],
targets: [{
allowAuthOverride: false,
config: {
databricks: {
cacheReadCost: 8.45,
cacheWriteCost: 3.62,
cacheWriteCostLists: [{
cost: 3.12,
ttl: "...my_ttl...",
}],
contextWindowFactors: [{
above: "...my_above...",
inputFactor: 4.94,
outputFactor: 9.43,
}],
embeddingsDimensions: 1316728274,
inputCost: 9.06,
maxTokens: 1585442569,
outputCost: 7.78,
serviceTierFactors: [{
factor: 6.01,
tier: "...my_tier...",
}],
temperature: 3.33,
topK: 896181225,
topP: 7.55,
upstreamUrl: "https://distant-antelope.com",
workspaceInstanceId: "...my_workspace_instance_id...",
},
},
name: "gpt-5-model",
provider: "azure-ai-se",
semanticDescription: "...my_semantic_description...",
weight: 100,
}],
},
});
import pulumi
import json
import pulumi_konnect as konnect
my_aigatewaymodel = konnect.AiGatewayModel("my_aigatewaymodel",
api={
"access": {
"acls": {
"allows": ["..."],
"denies": ["..."],
},
"auth_strategies": ["okta-ai-se"],
"identity_providers": ["okta-ai-se"],
},
"capabilities": ["files"],
"config": {
"balancer": {
"lowest_latency": {
"connect_timeout": 60000,
"fail_timeout": 10000,
"failover_criterias": ["http_429"],
"latency_strategy": "tpot",
"max_fails": 0,
"read_timeout": 60000,
"retries": 5,
"slots": 10000,
"write_timeout": 60000,
},
},
"logging": {
"payloads": False,
},
"max_request_body_size": 8388608,
"proxy": {
"auth": {
"password": "...my_password...",
"username": "...my_username...",
},
"http_proxy": {
"host": "...my_host...",
"port": 29747,
},
"https_proxy": {
"host": "...my_host...",
"port": 12764,
},
"no_proxy": "...my_no_proxy...",
"proxy_scheme": "http",
},
"response_streaming": "allow",
"route": {
"headers": {
"key": json.dumps("value"),
},
"hosts": ["foo.example.com"],
"https_redirect_status_code": 426,
"methods": ["..."],
"model": {
"body_param": "model",
"header_param": "x-model",
"path_param": "model_name",
"values": ["..."],
},
"paths": ["..."],
"preserve_host": False,
"protocols": ["..."],
"regex_priority": 0,
"request_buffering": True,
"response_buffering": True,
"strip_path": True,
"tags": ["..."],
},
},
"display_name": "My GPT 5 model",
"enabled": True,
"formats": [{
"type": "openai",
}],
"labels": {
"key": "value",
},
"managed_by": {
"key": "value",
},
"name": "my-gpt-5-model",
"policies": ["..."],
"targets": [{
"allow_auth_override": False,
"config": {
"xai": {
"cache_read_cost": 4.42,
"cache_write_cost": 3.9,
"cache_write_cost_lists": [{
"cost": 7.32,
"ttl": "...my_ttl...",
}],
"context_window_factors": [{
"above": "...my_above...",
"input_factor": 1.42,
"output_factor": 1.31,
}],
"embeddings_dimensions": 1556463673,
"input_cost": 3.7,
"max_tokens": 1227329724,
"output_cost": 6.56,
"service_tier_factors": [{
"factor": 8.57,
"tier": "...my_tier...",
}],
"temperature": 3.27,
"top_k": 483136424,
"top_p": 2.83,
"upstream_url": "https://baggy-trash.biz/",
},
},
"name": "gpt-5-model",
"provider": "azure-ai-se",
"semantic_description": "...my_semantic_description...",
"weight": 100,
}],
},
gateway_id="5f9fd312-a987-4628-b4c5-bb4f4fddd5f7",
model={
"access": {
"acls": {
"allows": ["..."],
"denies": ["..."],
},
"auth_strategies": ["okta-ai-se"],
"identity_providers": ["okta-ai-se"],
},
"capabilities": ["audio/speech"],
"config": {
"balancer": {
"semantic": {
"connect_timeout": 60000,
"embeddings": {
"allow_auth_override": False,
"config": {
"ollama": {
"upstream_url": "...my_upstream_url...",
},
},
"name": "...my_name...",
"provider": "azure-ai-se",
},
"fail_timeout": 10000,
"failover_criterias": ["http_502"],
"max_fails": 0,
"read_timeout": 60000,
"retries": 5,
"slots": 10000,
"vectordb": {
"pgvector": {
"database": "kong-pgvector",
"dimensions": 6,
"distance_metric": "euclidean",
"host": "127.0.0.1",
"password": "...my_password...",
"port": 5432,
"ssl": {
"cert": "...my_cert...",
"cert_key": "...my_cert_key...",
"enabled": True,
"required": True,
"verify": True,
"version": "tlsv1_2",
},
"threshold": 3.66,
"timeout": 5000,
"user": "postgres",
},
},
"write_timeout": 60000,
},
},
"logging": {
"payloads": False,
},
"max_request_body_size": 8388608,
"model": {
"name_header": True,
},
"proxy": {
"auth": {
"password": "...my_password...",
"username": "...my_username...",
},
"http_proxy": {
"host": "...my_host...",
"port": 30633,
},
"https_proxy": {
"host": "...my_host...",
"port": 29606,
},
"no_proxy": "...my_no_proxy...",
"proxy_scheme": "http",
},
"response_streaming": "allow",
"route": {
"headers": {
"key": json.dumps("value"),
},
"hosts": ["foo.example.com"],
"https_redirect_status_code": 426,
"methods": ["..."],
"model": {
"body_param": "model",
"header_param": "x-model",
"path_param": "model_name",
"values": ["..."],
},
"paths": ["..."],
"preserve_host": False,
"protocols": ["..."],
"regex_priority": 0,
"request_buffering": True,
"response_buffering": True,
"strip_path": True,
"tags": ["..."],
},
},
"display_name": "My GPT 5 model",
"enabled": True,
"formats": [{
"type": "openai",
}],
"labels": {
"key": "value",
},
"managed_by": {
"key": "value",
},
"name": "my-gpt-5-model",
"policies": ["..."],
"targets": [{
"allow_auth_override": False,
"config": {
"databricks": {
"cache_read_cost": 8.45,
"cache_write_cost": 3.62,
"cache_write_cost_lists": [{
"cost": 3.12,
"ttl": "...my_ttl...",
}],
"context_window_factors": [{
"above": "...my_above...",
"input_factor": 4.94,
"output_factor": 9.43,
}],
"embeddings_dimensions": 1316728274,
"input_cost": 9.06,
"max_tokens": 1585442569,
"output_cost": 7.78,
"service_tier_factors": [{
"factor": 6.01,
"tier": "...my_tier...",
}],
"temperature": 3.33,
"top_k": 896181225,
"top_p": 7.55,
"upstream_url": "https://distant-antelope.com",
"workspace_instance_id": "...my_workspace_instance_id...",
},
},
"name": "gpt-5-model",
"provider": "azure-ai-se",
"semantic_description": "...my_semantic_description...",
"weight": 100,
}],
})
package main
import (
"encoding/json"
"github.com/pulumi/pulumi-terraform-provider/sdks/go/konnect/v3/konnect"
"github.com/pulumi/pulumi/sdk/v3/go/pulumi"
)
func main() {
pulumi.Run(func(ctx *pulumi.Context) error {
tmpJSON0, err := json.Marshal("value")
if err != nil {
return err
}
json0 := string(tmpJSON0)
tmpJSON1, err := json.Marshal("value")
if err != nil {
return err
}
json1 := string(tmpJSON1)
_, err = konnect.NewAiGatewayModel(ctx, "my_aigatewaymodel", &konnect.AiGatewayModelArgs{
Api: &konnect.AiGatewayModelApiArgs{
Access: &konnect.AiGatewayModelApiAccessArgs{
Acls: &konnect.AiGatewayModelApiAccessAclsArgs{
Allows: pulumi.StringArray{
pulumi.String("..."),
},
Denies: pulumi.StringArray{
pulumi.String("..."),
},
},
AuthStrategies: pulumi.StringArray{
pulumi.String("okta-ai-se"),
},
IdentityProviders: pulumi.StringArray{
pulumi.String("okta-ai-se"),
},
},
Capabilities: pulumi.StringArray{
pulumi.String("files"),
},
Config: &konnect.AiGatewayModelApiConfigArgs{
Balancer: &konnect.AiGatewayModelApiConfigBalancerArgs{
LowestLatency: &konnect.AiGatewayModelApiConfigBalancerLowestLatencyArgs{
ConnectTimeout: pulumi.Float64(60000),
FailTimeout: pulumi.Float64(10000),
FailoverCriterias: pulumi.StringArray{
pulumi.String("http_429"),
},
LatencyStrategy: pulumi.String("tpot"),
MaxFails: pulumi.Float64(0),
ReadTimeout: pulumi.Float64(60000),
Retries: pulumi.Float64(5),
Slots: pulumi.Float64(10000),
WriteTimeout: pulumi.Float64(60000),
},
},
Logging: &konnect.AiGatewayModelApiConfigLoggingArgs{
Payloads: pulumi.Bool(false),
},
MaxRequestBodySize: pulumi.Float64(8388608),
Proxy: &konnect.AiGatewayModelApiConfigProxyArgs{
Auth: &konnect.AiGatewayModelApiConfigProxyAuthArgs{
Password: pulumi.String("...my_password..."),
Username: pulumi.String("...my_username..."),
},
HttpProxy: &konnect.AiGatewayModelApiConfigProxyHttpProxyArgs{
Host: pulumi.String("...my_host..."),
Port: pulumi.Float64(29747),
},
HttpsProxy: &konnect.AiGatewayModelApiConfigProxyHttpsProxyArgs{
Host: pulumi.String("...my_host..."),
Port: pulumi.Float64(12764),
},
NoProxy: pulumi.String("...my_no_proxy..."),
ProxyScheme: pulumi.String("http"),
},
ResponseStreaming: pulumi.String("allow"),
Route: &konnect.AiGatewayModelApiConfigRouteArgs{
Headers: pulumi.StringMap{
"key": pulumi.String(json0),
},
Hosts: pulumi.StringArray{
pulumi.String("foo.example.com"),
},
HttpsRedirectStatusCode: pulumi.Float64(426),
Methods: pulumi.StringArray{
pulumi.String("..."),
},
Model: &konnect.AiGatewayModelApiConfigRouteModelArgs{
BodyParam: pulumi.String("model"),
HeaderParam: pulumi.String("x-model"),
PathParam: pulumi.String("model_name"),
Values: pulumi.StringArray{
pulumi.String("..."),
},
},
Paths: pulumi.StringArray{
pulumi.String("..."),
},
PreserveHost: pulumi.Bool(false),
Protocols: pulumi.StringArray{
pulumi.String("..."),
},
RegexPriority: pulumi.Float64(0),
RequestBuffering: pulumi.Bool(true),
ResponseBuffering: pulumi.Bool(true),
StripPath: pulumi.Bool(true),
Tags: pulumi.StringArray{
pulumi.String("..."),
},
},
},
DisplayName: pulumi.String("My GPT 5 model"),
Enabled: pulumi.Bool(true),
Formats: konnect.AiGatewayModelApiFormatArray{
&konnect.AiGatewayModelApiFormatArgs{
Type: pulumi.String("openai"),
},
},
Labels: pulumi.StringMap{
"key": pulumi.String("value"),
},
ManagedBy: pulumi.StringMap{
"key": pulumi.String("value"),
},
Name: pulumi.String("my-gpt-5-model"),
Policies: pulumi.StringArray{
pulumi.String("..."),
},
Targets: konnect.AiGatewayModelApiTargetArray{
&konnect.AiGatewayModelApiTargetArgs{
AllowAuthOverride: pulumi.Bool(false),
Config: &konnect.AiGatewayModelApiTargetConfigArgs{
Xai: &konnect.AiGatewayModelApiTargetConfigXaiArgs{
CacheReadCost: pulumi.Float64(4.42),
CacheWriteCost: pulumi.Float64(3.9),
CacheWriteCostLists: konnect.AiGatewayModelApiTargetConfigXaiCacheWriteCostListArray{
&konnect.AiGatewayModelApiTargetConfigXaiCacheWriteCostListArgs{
Cost: pulumi.Float64(7.32),
Ttl: pulumi.String("...my_ttl..."),
},
},
ContextWindowFactors: konnect.AiGatewayModelApiTargetConfigXaiContextWindowFactorArray{
&konnect.AiGatewayModelApiTargetConfigXaiContextWindowFactorArgs{
Above: pulumi.String("...my_above..."),
InputFactor: pulumi.Float64(1.42),
OutputFactor: pulumi.Float64(1.31),
},
},
EmbeddingsDimensions: pulumi.Float64(1556463673),
InputCost: pulumi.Float64(3.7),
MaxTokens: pulumi.Float64(1227329724),
OutputCost: pulumi.Float64(6.56),
ServiceTierFactors: konnect.AiGatewayModelApiTargetConfigXaiServiceTierFactorArray{
&konnect.AiGatewayModelApiTargetConfigXaiServiceTierFactorArgs{
Factor: pulumi.Float64(8.57),
Tier: pulumi.String("...my_tier..."),
},
},
Temperature: pulumi.Float64(3.27),
TopK: pulumi.Float64(483136424),
TopP: pulumi.Float64(2.83),
UpstreamUrl: pulumi.String("https://baggy-trash.biz/"),
},
},
Name: pulumi.String("gpt-5-model"),
Provider: pulumi.String("azure-ai-se"),
SemanticDescription: pulumi.String("...my_semantic_description..."),
Weight: pulumi.Float64(100),
},
},
},
GatewayId: pulumi.String("5f9fd312-a987-4628-b4c5-bb4f4fddd5f7"),
Model: &konnect.AiGatewayModelModelArgs{
Access: &konnect.AiGatewayModelModelAccessArgs{
Acls: &konnect.AiGatewayModelModelAccessAclsArgs{
Allows: pulumi.StringArray{
pulumi.String("..."),
},
Denies: pulumi.StringArray{
pulumi.String("..."),
},
},
AuthStrategies: pulumi.StringArray{
pulumi.String("okta-ai-se"),
},
IdentityProviders: pulumi.StringArray{
pulumi.String("okta-ai-se"),
},
},
Capabilities: pulumi.StringArray{
pulumi.String("audio/speech"),
},
Config: &konnect.AiGatewayModelModelConfigArgs{
Balancer: &konnect.AiGatewayModelModelConfigBalancerArgs{
Semantic: &konnect.AiGatewayModelModelConfigBalancerSemanticArgs{
ConnectTimeout: pulumi.Float64(60000),
Embeddings: &konnect.AiGatewayModelModelConfigBalancerSemanticEmbeddingsArgs{
AllowAuthOverride: pulumi.Bool(false),
Config: &konnect.AiGatewayModelModelConfigBalancerSemanticEmbeddingsConfigArgs{
Ollama: &konnect.AiGatewayModelModelConfigBalancerSemanticEmbeddingsConfigOllamaArgs{
UpstreamUrl: pulumi.String("...my_upstream_url..."),
},
},
Name: pulumi.String("...my_name..."),
Provider: pulumi.String("azure-ai-se"),
},
FailTimeout: pulumi.Float64(10000),
FailoverCriterias: pulumi.StringArray{
pulumi.String("http_502"),
},
MaxFails: pulumi.Float64(0),
ReadTimeout: pulumi.Float64(60000),
Retries: pulumi.Float64(5),
Slots: pulumi.Float64(10000),
Vectordb: &konnect.AiGatewayModelModelConfigBalancerSemanticVectordbArgs{
Pgvector: &konnect.AiGatewayModelModelConfigBalancerSemanticVectordbPgvectorArgs{
Database: pulumi.String("kong-pgvector"),
Dimensions: pulumi.Float64(6),
DistanceMetric: pulumi.String("euclidean"),
Host: pulumi.String("127.0.0.1"),
Password: pulumi.String("...my_password..."),
Port: pulumi.Float64(5432),
Ssl: &konnect.AiGatewayModelModelConfigBalancerSemanticVectordbPgvectorSslArgs{
Cert: pulumi.String("...my_cert..."),
CertKey: pulumi.String("...my_cert_key..."),
Enabled: pulumi.Bool(true),
Required: pulumi.Bool(true),
Verify: pulumi.Bool(true),
Version: pulumi.String("tlsv1_2"),
},
Threshold: pulumi.Float64(3.66),
Timeout: pulumi.Float64(5000),
User: pulumi.String("postgres"),
},
},
WriteTimeout: pulumi.Float64(60000),
},
},
Logging: &konnect.AiGatewayModelModelConfigLoggingArgs{
Payloads: pulumi.Bool(false),
},
MaxRequestBodySize: pulumi.Float64(8388608),
Model: &konnect.AiGatewayModelModelConfigModelArgs{
NameHeader: pulumi.Bool(true),
},
Proxy: &konnect.AiGatewayModelModelConfigProxyArgs{
Auth: &konnect.AiGatewayModelModelConfigProxyAuthArgs{
Password: pulumi.String("...my_password..."),
Username: pulumi.String("...my_username..."),
},
HttpProxy: &konnect.AiGatewayModelModelConfigProxyHttpProxyArgs{
Host: pulumi.String("...my_host..."),
Port: pulumi.Float64(30633),
},
HttpsProxy: &konnect.AiGatewayModelModelConfigProxyHttpsProxyArgs{
Host: pulumi.String("...my_host..."),
Port: pulumi.Float64(29606),
},
NoProxy: pulumi.String("...my_no_proxy..."),
ProxyScheme: pulumi.String("http"),
},
ResponseStreaming: pulumi.String("allow"),
Route: &konnect.AiGatewayModelModelConfigRouteArgs{
Headers: pulumi.StringMap{
"key": pulumi.String(json1),
},
Hosts: pulumi.StringArray{
pulumi.String("foo.example.com"),
},
HttpsRedirectStatusCode: pulumi.Float64(426),
Methods: pulumi.StringArray{
pulumi.String("..."),
},
Model: &konnect.AiGatewayModelModelConfigRouteModelArgs{
BodyParam: pulumi.String("model"),
HeaderParam: pulumi.String("x-model"),
PathParam: pulumi.String("model_name"),
Values: pulumi.StringArray{
pulumi.String("..."),
},
},
Paths: pulumi.StringArray{
pulumi.String("..."),
},
PreserveHost: pulumi.Bool(false),
Protocols: pulumi.StringArray{
pulumi.String("..."),
},
RegexPriority: pulumi.Float64(0),
RequestBuffering: pulumi.Bool(true),
ResponseBuffering: pulumi.Bool(true),
StripPath: pulumi.Bool(true),
Tags: pulumi.StringArray{
pulumi.String("..."),
},
},
},
DisplayName: pulumi.String("My GPT 5 model"),
Enabled: pulumi.Bool(true),
Formats: konnect.AiGatewayModelModelFormatArray{
&konnect.AiGatewayModelModelFormatArgs{
Type: pulumi.String("openai"),
},
},
Labels: pulumi.StringMap{
"key": pulumi.String("value"),
},
ManagedBy: pulumi.StringMap{
"key": pulumi.String("value"),
},
Name: pulumi.String("my-gpt-5-model"),
Policies: pulumi.StringArray{
pulumi.String("..."),
},
Targets: konnect.AiGatewayModelModelTargetArray{
&konnect.AiGatewayModelModelTargetArgs{
AllowAuthOverride: pulumi.Bool(false),
Config: &konnect.AiGatewayModelModelTargetConfigArgs{
Databricks: &konnect.AiGatewayModelModelTargetConfigDatabricksArgs{
CacheReadCost: pulumi.Float64(8.45),
CacheWriteCost: pulumi.Float64(3.62),
CacheWriteCostLists: konnect.AiGatewayModelModelTargetConfigDatabricksCacheWriteCostListArray{
&konnect.AiGatewayModelModelTargetConfigDatabricksCacheWriteCostListArgs{
Cost: pulumi.Float64(3.12),
Ttl: pulumi.String("...my_ttl..."),
},
},
ContextWindowFactors: konnect.AiGatewayModelModelTargetConfigDatabricksContextWindowFactorArray{
&konnect.AiGatewayModelModelTargetConfigDatabricksContextWindowFactorArgs{
Above: pulumi.String("...my_above..."),
InputFactor: pulumi.Float64(4.94),
OutputFactor: pulumi.Float64(9.43),
},
},
EmbeddingsDimensions: pulumi.Float64(1316728274),
InputCost: pulumi.Float64(9.06),
MaxTokens: pulumi.Float64(1585442569),
OutputCost: pulumi.Float64(7.78),
ServiceTierFactors: konnect.AiGatewayModelModelTargetConfigDatabricksServiceTierFactorArray{
&konnect.AiGatewayModelModelTargetConfigDatabricksServiceTierFactorArgs{
Factor: pulumi.Float64(6.01),
Tier: pulumi.String("...my_tier..."),
},
},
Temperature: pulumi.Float64(3.33),
TopK: pulumi.Float64(896181225),
TopP: pulumi.Float64(7.55),
UpstreamUrl: pulumi.String("https://distant-antelope.com"),
WorkspaceInstanceId: pulumi.String("...my_workspace_instance_id..."),
},
},
Name: pulumi.String("gpt-5-model"),
Provider: pulumi.String("azure-ai-se"),
SemanticDescription: pulumi.String("...my_semantic_description..."),
Weight: pulumi.Float64(100),
},
},
},
})
if err != nil {
return err
}
return nil
})
}
using System.Collections.Generic;
using System.Linq;
using System.Text.Json;
using Pulumi;
using Konnect = Pulumi.Konnect;
return await Deployment.RunAsync(() =>
{
var myAigatewaymodel = new Konnect.AiGatewayModel("my_aigatewaymodel", new()
{
Api = new Konnect.Inputs.AiGatewayModelApiArgs
{
Access = new Konnect.Inputs.AiGatewayModelApiAccessArgs
{
Acls = new Konnect.Inputs.AiGatewayModelApiAccessAclsArgs
{
Allows = new[]
{
"...",
},
Denies = new[]
{
"...",
},
},
AuthStrategies = new[]
{
"okta-ai-se",
},
IdentityProviders = new[]
{
"okta-ai-se",
},
},
Capabilities = new[]
{
"files",
},
Config = new Konnect.Inputs.AiGatewayModelApiConfigArgs
{
Balancer = new Konnect.Inputs.AiGatewayModelApiConfigBalancerArgs
{
LowestLatency = new Konnect.Inputs.AiGatewayModelApiConfigBalancerLowestLatencyArgs
{
ConnectTimeout = 60000,
FailTimeout = 10000,
FailoverCriterias = new[]
{
"http_429",
},
LatencyStrategy = "tpot",
MaxFails = 0,
ReadTimeout = 60000,
Retries = 5,
Slots = 10000,
WriteTimeout = 60000,
},
},
Logging = new Konnect.Inputs.AiGatewayModelApiConfigLoggingArgs
{
Payloads = false,
},
MaxRequestBodySize = 8388608,
Proxy = new Konnect.Inputs.AiGatewayModelApiConfigProxyArgs
{
Auth = new Konnect.Inputs.AiGatewayModelApiConfigProxyAuthArgs
{
Password = "...my_password...",
Username = "...my_username...",
},
HttpProxy = new Konnect.Inputs.AiGatewayModelApiConfigProxyHttpProxyArgs
{
Host = "...my_host...",
Port = 29747,
},
HttpsProxy = new Konnect.Inputs.AiGatewayModelApiConfigProxyHttpsProxyArgs
{
Host = "...my_host...",
Port = 12764,
},
NoProxy = "...my_no_proxy...",
ProxyScheme = "http",
},
ResponseStreaming = "allow",
Route = new Konnect.Inputs.AiGatewayModelApiConfigRouteArgs
{
Headers =
{
{ "key", JsonSerializer.Serialize("value") },
},
Hosts = new[]
{
"foo.example.com",
},
HttpsRedirectStatusCode = 426,
Methods = new[]
{
"...",
},
Model = new Konnect.Inputs.AiGatewayModelApiConfigRouteModelArgs
{
BodyParam = "model",
HeaderParam = "x-model",
PathParam = "model_name",
Values = new[]
{
"...",
},
},
Paths = new[]
{
"...",
},
PreserveHost = false,
Protocols = new[]
{
"...",
},
RegexPriority = 0,
RequestBuffering = true,
ResponseBuffering = true,
StripPath = true,
Tags = new[]
{
"...",
},
},
},
DisplayName = "My GPT 5 model",
Enabled = true,
Formats = new[]
{
new Konnect.Inputs.AiGatewayModelApiFormatArgs
{
Type = "openai",
},
},
Labels =
{
{ "key", "value" },
},
ManagedBy =
{
{ "key", "value" },
},
Name = "my-gpt-5-model",
Policies = new[]
{
"...",
},
Targets = new[]
{
new Konnect.Inputs.AiGatewayModelApiTargetArgs
{
AllowAuthOverride = false,
Config = new Konnect.Inputs.AiGatewayModelApiTargetConfigArgs
{
Xai = new Konnect.Inputs.AiGatewayModelApiTargetConfigXaiArgs
{
CacheReadCost = 4.42,
CacheWriteCost = 3.9,
CacheWriteCostLists = new[]
{
new Konnect.Inputs.AiGatewayModelApiTargetConfigXaiCacheWriteCostListArgs
{
Cost = 7.32,
Ttl = "...my_ttl...",
},
},
ContextWindowFactors = new[]
{
new Konnect.Inputs.AiGatewayModelApiTargetConfigXaiContextWindowFactorArgs
{
Above = "...my_above...",
InputFactor = 1.42,
OutputFactor = 1.31,
},
},
EmbeddingsDimensions = 1556463673,
InputCost = 3.7,
MaxTokens = 1227329724,
OutputCost = 6.56,
ServiceTierFactors = new[]
{
new Konnect.Inputs.AiGatewayModelApiTargetConfigXaiServiceTierFactorArgs
{
Factor = 8.57,
Tier = "...my_tier...",
},
},
Temperature = 3.27,
TopK = 483136424,
TopP = 2.83,
UpstreamUrl = "https://baggy-trash.biz/",
},
},
Name = "gpt-5-model",
Provider = "azure-ai-se",
SemanticDescription = "...my_semantic_description...",
Weight = 100,
},
},
},
GatewayId = "5f9fd312-a987-4628-b4c5-bb4f4fddd5f7",
Model = new Konnect.Inputs.AiGatewayModelModelArgs
{
Access = new Konnect.Inputs.AiGatewayModelModelAccessArgs
{
Acls = new Konnect.Inputs.AiGatewayModelModelAccessAclsArgs
{
Allows = new[]
{
"...",
},
Denies = new[]
{
"...",
},
},
AuthStrategies = new[]
{
"okta-ai-se",
},
IdentityProviders = new[]
{
"okta-ai-se",
},
},
Capabilities = new[]
{
"audio/speech",
},
Config = new Konnect.Inputs.AiGatewayModelModelConfigArgs
{
Balancer = new Konnect.Inputs.AiGatewayModelModelConfigBalancerArgs
{
Semantic = new Konnect.Inputs.AiGatewayModelModelConfigBalancerSemanticArgs
{
ConnectTimeout = 60000,
Embeddings = new Konnect.Inputs.AiGatewayModelModelConfigBalancerSemanticEmbeddingsArgs
{
AllowAuthOverride = false,
Config = new Konnect.Inputs.AiGatewayModelModelConfigBalancerSemanticEmbeddingsConfigArgs
{
Ollama = new Konnect.Inputs.AiGatewayModelModelConfigBalancerSemanticEmbeddingsConfigOllamaArgs
{
UpstreamUrl = "...my_upstream_url...",
},
},
Name = "...my_name...",
Provider = "azure-ai-se",
},
FailTimeout = 10000,
FailoverCriterias = new[]
{
"http_502",
},
MaxFails = 0,
ReadTimeout = 60000,
Retries = 5,
Slots = 10000,
Vectordb = new Konnect.Inputs.AiGatewayModelModelConfigBalancerSemanticVectordbArgs
{
Pgvector = new Konnect.Inputs.AiGatewayModelModelConfigBalancerSemanticVectordbPgvectorArgs
{
Database = "kong-pgvector",
Dimensions = 6,
DistanceMetric = "euclidean",
Host = "127.0.0.1",
Password = "...my_password...",
Port = 5432,
Ssl = new Konnect.Inputs.AiGatewayModelModelConfigBalancerSemanticVectordbPgvectorSslArgs
{
Cert = "...my_cert...",
CertKey = "...my_cert_key...",
Enabled = true,
Required = true,
Verify = true,
Version = "tlsv1_2",
},
Threshold = 3.66,
Timeout = 5000,
User = "postgres",
},
},
WriteTimeout = 60000,
},
},
Logging = new Konnect.Inputs.AiGatewayModelModelConfigLoggingArgs
{
Payloads = false,
},
MaxRequestBodySize = 8388608,
Model = new Konnect.Inputs.AiGatewayModelModelConfigModelArgs
{
NameHeader = true,
},
Proxy = new Konnect.Inputs.AiGatewayModelModelConfigProxyArgs
{
Auth = new Konnect.Inputs.AiGatewayModelModelConfigProxyAuthArgs
{
Password = "...my_password...",
Username = "...my_username...",
},
HttpProxy = new Konnect.Inputs.AiGatewayModelModelConfigProxyHttpProxyArgs
{
Host = "...my_host...",
Port = 30633,
},
HttpsProxy = new Konnect.Inputs.AiGatewayModelModelConfigProxyHttpsProxyArgs
{
Host = "...my_host...",
Port = 29606,
},
NoProxy = "...my_no_proxy...",
ProxyScheme = "http",
},
ResponseStreaming = "allow",
Route = new Konnect.Inputs.AiGatewayModelModelConfigRouteArgs
{
Headers =
{
{ "key", JsonSerializer.Serialize("value") },
},
Hosts = new[]
{
"foo.example.com",
},
HttpsRedirectStatusCode = 426,
Methods = new[]
{
"...",
},
Model = new Konnect.Inputs.AiGatewayModelModelConfigRouteModelArgs
{
BodyParam = "model",
HeaderParam = "x-model",
PathParam = "model_name",
Values = new[]
{
"...",
},
},
Paths = new[]
{
"...",
},
PreserveHost = false,
Protocols = new[]
{
"...",
},
RegexPriority = 0,
RequestBuffering = true,
ResponseBuffering = true,
StripPath = true,
Tags = new[]
{
"...",
},
},
},
DisplayName = "My GPT 5 model",
Enabled = true,
Formats = new[]
{
new Konnect.Inputs.AiGatewayModelModelFormatArgs
{
Type = "openai",
},
},
Labels =
{
{ "key", "value" },
},
ManagedBy =
{
{ "key", "value" },
},
Name = "my-gpt-5-model",
Policies = new[]
{
"...",
},
Targets = new[]
{
new Konnect.Inputs.AiGatewayModelModelTargetArgs
{
AllowAuthOverride = false,
Config = new Konnect.Inputs.AiGatewayModelModelTargetConfigArgs
{
Databricks = new Konnect.Inputs.AiGatewayModelModelTargetConfigDatabricksArgs
{
CacheReadCost = 8.45,
CacheWriteCost = 3.62,
CacheWriteCostLists = new[]
{
new Konnect.Inputs.AiGatewayModelModelTargetConfigDatabricksCacheWriteCostListArgs
{
Cost = 3.12,
Ttl = "...my_ttl...",
},
},
ContextWindowFactors = new[]
{
new Konnect.Inputs.AiGatewayModelModelTargetConfigDatabricksContextWindowFactorArgs
{
Above = "...my_above...",
InputFactor = 4.94,
OutputFactor = 9.43,
},
},
EmbeddingsDimensions = 1316728274,
InputCost = 9.06,
MaxTokens = 1585442569,
OutputCost = 7.78,
ServiceTierFactors = new[]
{
new Konnect.Inputs.AiGatewayModelModelTargetConfigDatabricksServiceTierFactorArgs
{
Factor = 6.01,
Tier = "...my_tier...",
},
},
Temperature = 3.33,
TopK = 896181225,
TopP = 7.55,
UpstreamUrl = "https://distant-antelope.com",
WorkspaceInstanceId = "...my_workspace_instance_id...",
},
},
Name = "gpt-5-model",
Provider = "azure-ai-se",
SemanticDescription = "...my_semantic_description...",
Weight = 100,
},
},
},
});
});
package generated_program;
import com.pulumi.Context;
import com.pulumi.Pulumi;
import com.pulumi.core.Output;
import com.pulumi.konnect.AiGatewayModel;
import com.pulumi.konnect.AiGatewayModelArgs;
import com.pulumi.konnect.inputs.AiGatewayModelApiArgs;
import com.pulumi.konnect.inputs.AiGatewayModelApiAccessArgs;
import com.pulumi.konnect.inputs.AiGatewayModelApiAccessAclsArgs;
import com.pulumi.konnect.inputs.AiGatewayModelApiConfigArgs;
import com.pulumi.konnect.inputs.AiGatewayModelApiConfigBalancerArgs;
import com.pulumi.konnect.inputs.AiGatewayModelApiConfigBalancerLowestLatencyArgs;
import com.pulumi.konnect.inputs.AiGatewayModelApiConfigLoggingArgs;
import com.pulumi.konnect.inputs.AiGatewayModelApiConfigProxyArgs;
import com.pulumi.konnect.inputs.AiGatewayModelApiConfigProxyAuthArgs;
import com.pulumi.konnect.inputs.AiGatewayModelApiConfigProxyHttpProxyArgs;
import com.pulumi.konnect.inputs.AiGatewayModelApiConfigProxyHttpsProxyArgs;
import com.pulumi.konnect.inputs.AiGatewayModelApiConfigRouteArgs;
import com.pulumi.konnect.inputs.AiGatewayModelApiConfigRouteModelArgs;
import com.pulumi.konnect.inputs.AiGatewayModelModelArgs;
import com.pulumi.konnect.inputs.AiGatewayModelModelAccessArgs;
import com.pulumi.konnect.inputs.AiGatewayModelModelAccessAclsArgs;
import com.pulumi.konnect.inputs.AiGatewayModelModelConfigArgs;
import com.pulumi.konnect.inputs.AiGatewayModelModelConfigBalancerArgs;
import com.pulumi.konnect.inputs.AiGatewayModelModelConfigBalancerSemanticArgs;
import com.pulumi.konnect.inputs.AiGatewayModelModelConfigBalancerSemanticEmbeddingsArgs;
import com.pulumi.konnect.inputs.AiGatewayModelModelConfigBalancerSemanticEmbeddingsConfigArgs;
import com.pulumi.konnect.inputs.AiGatewayModelModelConfigBalancerSemanticEmbeddingsConfigOllamaArgs;
import com.pulumi.konnect.inputs.AiGatewayModelModelConfigBalancerSemanticVectordbArgs;
import com.pulumi.konnect.inputs.AiGatewayModelModelConfigBalancerSemanticVectordbPgvectorArgs;
import com.pulumi.konnect.inputs.AiGatewayModelModelConfigBalancerSemanticVectordbPgvectorSslArgs;
import com.pulumi.konnect.inputs.AiGatewayModelModelConfigLoggingArgs;
import com.pulumi.konnect.inputs.AiGatewayModelModelConfigModelArgs;
import com.pulumi.konnect.inputs.AiGatewayModelModelConfigProxyArgs;
import com.pulumi.konnect.inputs.AiGatewayModelModelConfigProxyAuthArgs;
import com.pulumi.konnect.inputs.AiGatewayModelModelConfigProxyHttpProxyArgs;
import com.pulumi.konnect.inputs.AiGatewayModelModelConfigProxyHttpsProxyArgs;
import com.pulumi.konnect.inputs.AiGatewayModelModelConfigRouteArgs;
import com.pulumi.konnect.inputs.AiGatewayModelModelConfigRouteModelArgs;
import static com.pulumi.codegen.internal.Serialization.*;
import java.util.List;
import java.util.ArrayList;
import java.util.Map;
import java.io.File;
import java.nio.file.Files;
import java.nio.file.Paths;
public class App {
public static void main(String[] args) {
Pulumi.run(App::stack);
}
public static void stack(Context ctx) {
var myAigatewaymodel = new AiGatewayModel("myAigatewaymodel", AiGatewayModelArgs.builder()
.api(AiGatewayModelApiArgs.builder()
.access(AiGatewayModelApiAccessArgs.builder()
.acls(AiGatewayModelApiAccessAclsArgs.builder()
.allows("...")
.denies("...")
.build())
.authStrategies("okta-ai-se")
.identityProviders("okta-ai-se")
.build())
.capabilities("files")
.config(AiGatewayModelApiConfigArgs.builder()
.balancer(AiGatewayModelApiConfigBalancerArgs.builder()
.lowestLatency(AiGatewayModelApiConfigBalancerLowestLatencyArgs.builder()
.connectTimeout(60000.0)
.failTimeout(10000.0)
.failoverCriterias("http_429")
.latencyStrategy("tpot")
.maxFails(0.0)
.readTimeout(60000.0)
.retries(5.0)
.slots(10000.0)
.writeTimeout(60000.0)
.build())
.build())
.logging(AiGatewayModelApiConfigLoggingArgs.builder()
.payloads(false)
.build())
.maxRequestBodySize(8388608.0)
.proxy(AiGatewayModelApiConfigProxyArgs.builder()
.auth(AiGatewayModelApiConfigProxyAuthArgs.builder()
.password("...my_password...")
.username("...my_username...")
.build())
.httpProxy(AiGatewayModelApiConfigProxyHttpProxyArgs.builder()
.host("...my_host...")
.port(29747.0)
.build())
.httpsProxy(AiGatewayModelApiConfigProxyHttpsProxyArgs.builder()
.host("...my_host...")
.port(12764.0)
.build())
.noProxy("...my_no_proxy...")
.proxyScheme("http")
.build())
.responseStreaming("allow")
.route(AiGatewayModelApiConfigRouteArgs.builder()
.headers(Map.of("key", serializeJson(
"value")))
.hosts("foo.example.com")
.httpsRedirectStatusCode(426.0)
.methods("...")
.model(AiGatewayModelApiConfigRouteModelArgs.builder()
.bodyParam("model")
.headerParam("x-model")
.pathParam("model_name")
.values("...")
.build())
.paths("...")
.preserveHost(false)
.protocols("...")
.regexPriority(0.0)
.requestBuffering(true)
.responseBuffering(true)
.stripPath(true)
.tags("...")
.build())
.build())
.displayName("My GPT 5 model")
.enabled(true)
.formats(AiGatewayModelApiFormatArgs.builder()
.type("openai")
.build())
.labels(Map.of("key", "value"))
.managedBy(Map.of("key", "value"))
.name("my-gpt-5-model")
.policies("...")
.targets(AiGatewayModelApiTargetArgs.builder()
.allowAuthOverride(false)
.config(AiGatewayModelApiTargetConfigArgs.builder()
.xai(AiGatewayModelApiTargetConfigXaiArgs.builder()
.cacheReadCost(4.42)
.cacheWriteCost(3.9)
.cacheWriteCostLists(AiGatewayModelApiTargetConfigXaiCacheWriteCostListArgs.builder()
.cost(7.32)
.ttl("...my_ttl...")
.build())
.contextWindowFactors(AiGatewayModelApiTargetConfigXaiContextWindowFactorArgs.builder()
.above("...my_above...")
.inputFactor(1.42)
.outputFactor(1.31)
.build())
.embeddingsDimensions(1556463673.0)
.inputCost(3.7)
.maxTokens(1227329724.0)
.outputCost(6.56)
.serviceTierFactors(AiGatewayModelApiTargetConfigXaiServiceTierFactorArgs.builder()
.factor(8.57)
.tier("...my_tier...")
.build())
.temperature(3.27)
.topK(483136424.0)
.topP(2.83)
.upstreamUrl("https://baggy-trash.biz/")
.build())
.build())
.name("gpt-5-model")
.provider("azure-ai-se")
.semanticDescription("...my_semantic_description...")
.weight(100.0)
.build())
.build())
.gatewayId("5f9fd312-a987-4628-b4c5-bb4f4fddd5f7")
.model(AiGatewayModelModelArgs.builder()
.access(AiGatewayModelModelAccessArgs.builder()
.acls(AiGatewayModelModelAccessAclsArgs.builder()
.allows("...")
.denies("...")
.build())
.authStrategies("okta-ai-se")
.identityProviders("okta-ai-se")
.build())
.capabilities("audio/speech")
.config(AiGatewayModelModelConfigArgs.builder()
.balancer(AiGatewayModelModelConfigBalancerArgs.builder()
.semantic(AiGatewayModelModelConfigBalancerSemanticArgs.builder()
.connectTimeout(60000.0)
.embeddings(AiGatewayModelModelConfigBalancerSemanticEmbeddingsArgs.builder()
.allowAuthOverride(false)
.config(AiGatewayModelModelConfigBalancerSemanticEmbeddingsConfigArgs.builder()
.ollama(AiGatewayModelModelConfigBalancerSemanticEmbeddingsConfigOllamaArgs.builder()
.upstreamUrl("...my_upstream_url...")
.build())
.build())
.name("...my_name...")
.provider("azure-ai-se")
.build())
.failTimeout(10000.0)
.failoverCriterias("http_502")
.maxFails(0.0)
.readTimeout(60000.0)
.retries(5.0)
.slots(10000.0)
.vectordb(AiGatewayModelModelConfigBalancerSemanticVectordbArgs.builder()
.pgvector(AiGatewayModelModelConfigBalancerSemanticVectordbPgvectorArgs.builder()
.database("kong-pgvector")
.dimensions(6.0)
.distanceMetric("euclidean")
.host("127.0.0.1")
.password("...my_password...")
.port(5432.0)
.ssl(AiGatewayModelModelConfigBalancerSemanticVectordbPgvectorSslArgs.builder()
.cert("...my_cert...")
.certKey("...my_cert_key...")
.enabled(true)
.required(true)
.verify(true)
.version("tlsv1_2")
.build())
.threshold(3.66)
.timeout(5000.0)
.user("postgres")
.build())
.build())
.writeTimeout(60000.0)
.build())
.build())
.logging(AiGatewayModelModelConfigLoggingArgs.builder()
.payloads(false)
.build())
.maxRequestBodySize(8388608.0)
.model(AiGatewayModelModelConfigModelArgs.builder()
.nameHeader(true)
.build())
.proxy(AiGatewayModelModelConfigProxyArgs.builder()
.auth(AiGatewayModelModelConfigProxyAuthArgs.builder()
.password("...my_password...")
.username("...my_username...")
.build())
.httpProxy(AiGatewayModelModelConfigProxyHttpProxyArgs.builder()
.host("...my_host...")
.port(30633.0)
.build())
.httpsProxy(AiGatewayModelModelConfigProxyHttpsProxyArgs.builder()
.host("...my_host...")
.port(29606.0)
.build())
.noProxy("...my_no_proxy...")
.proxyScheme("http")
.build())
.responseStreaming("allow")
.route(AiGatewayModelModelConfigRouteArgs.builder()
.headers(Map.of("key", serializeJson(
"value")))
.hosts("foo.example.com")
.httpsRedirectStatusCode(426.0)
.methods("...")
.model(AiGatewayModelModelConfigRouteModelArgs.builder()
.bodyParam("model")
.headerParam("x-model")
.pathParam("model_name")
.values("...")
.build())
.paths("...")
.preserveHost(false)
.protocols("...")
.regexPriority(0.0)
.requestBuffering(true)
.responseBuffering(true)
.stripPath(true)
.tags("...")
.build())
.build())
.displayName("My GPT 5 model")
.enabled(true)
.formats(AiGatewayModelModelFormatArgs.builder()
.type("openai")
.build())
.labels(Map.of("key", "value"))
.managedBy(Map.of("key", "value"))
.name("my-gpt-5-model")
.policies("...")
.targets(AiGatewayModelModelTargetArgs.builder()
.allowAuthOverride(false)
.config(AiGatewayModelModelTargetConfigArgs.builder()
.databricks(AiGatewayModelModelTargetConfigDatabricksArgs.builder()
.cacheReadCost(8.45)
.cacheWriteCost(3.62)
.cacheWriteCostLists(AiGatewayModelModelTargetConfigDatabricksCacheWriteCostListArgs.builder()
.cost(3.12)
.ttl("...my_ttl...")
.build())
.contextWindowFactors(AiGatewayModelModelTargetConfigDatabricksContextWindowFactorArgs.builder()
.above("...my_above...")
.inputFactor(4.94)
.outputFactor(9.43)
.build())
.embeddingsDimensions(1316728274.0)
.inputCost(9.06)
.maxTokens(1585442569.0)
.outputCost(7.78)
.serviceTierFactors(AiGatewayModelModelTargetConfigDatabricksServiceTierFactorArgs.builder()
.factor(6.01)
.tier("...my_tier...")
.build())
.temperature(3.33)
.topK(896181225.0)
.topP(7.55)
.upstreamUrl("https://distant-antelope.com")
.workspaceInstanceId("...my_workspace_instance_id...")
.build())
.build())
.name("gpt-5-model")
.provider("azure-ai-se")
.semanticDescription("...my_semantic_description...")
.weight(100.0)
.build())
.build())
.build());
}
}
resources:
myAigatewaymodel:
type: konnect:AiGatewayModel
name: my_aigatewaymodel
properties:
api:
access:
acls:
allows:
- '...'
denies:
- '...'
authStrategies:
- okta-ai-se
identityProviders:
- okta-ai-se
capabilities:
- files
config:
balancer:
lowestLatency:
connectTimeout: 60000
failTimeout: 10000
failoverCriterias:
- http_429
latencyStrategy: tpot
maxFails: 0
readTimeout: 60000
retries: 5
slots: 10000
writeTimeout: 60000
logging:
payloads: false
maxRequestBodySize: 8.388608e+06
proxy:
auth:
password: '...my_password...'
username: '...my_username...'
httpProxy:
host: '...my_host...'
port: 29747
httpsProxy:
host: '...my_host...'
port: 12764
noProxy: '...my_no_proxy...'
proxyScheme: http
responseStreaming: allow
route:
headers:
key:
fn::toJSON: value
hosts:
- foo.example.com
httpsRedirectStatusCode: 426
methods:
- '...'
model:
bodyParam: model
headerParam: x-model
pathParam: model_name
values:
- '...'
paths:
- '...'
preserveHost: false
protocols:
- '...'
regexPriority: 0
requestBuffering: true
responseBuffering: true
stripPath: true
tags:
- '...'
displayName: My GPT 5 model
enabled: true
formats:
- type: openai
labels:
key: value
managedBy:
key: value
name: my-gpt-5-model
policies:
- '...'
targets:
- allowAuthOverride: false
config:
xai:
cacheReadCost: 4.42
cacheWriteCost: 3.9
cacheWriteCostLists:
- cost: 7.32
ttl: '...my_ttl...'
contextWindowFactors:
- above: '...my_above...'
inputFactor: 1.42
outputFactor: 1.31
embeddingsDimensions: 1.556463673e+09
inputCost: 3.7
maxTokens: 1.227329724e+09
outputCost: 6.56
serviceTierFactors:
- factor: 8.57
tier: '...my_tier...'
temperature: 3.27
topK: 4.83136424e+08
topP: 2.83
upstreamUrl: https://baggy-trash.biz/
name: gpt-5-model
provider: azure-ai-se
semanticDescription: '...my_semantic_description...'
weight: 100
gatewayId: 5f9fd312-a987-4628-b4c5-bb4f4fddd5f7
model:
access:
acls:
allows:
- '...'
denies:
- '...'
authStrategies:
- okta-ai-se
identityProviders:
- okta-ai-se
capabilities:
- audio/speech
config:
balancer:
semantic:
connectTimeout: 60000
embeddings:
allowAuthOverride: false
config:
ollama:
upstreamUrl: '...my_upstream_url...'
name: '...my_name...'
provider: azure-ai-se
failTimeout: 10000
failoverCriterias:
- http_502
maxFails: 0
readTimeout: 60000
retries: 5
slots: 10000
vectordb:
pgvector:
database: kong-pgvector
dimensions: 6
distanceMetric: euclidean
host: 127.0.0.1
password: '...my_password...'
port: 5432
ssl:
cert: '...my_cert...'
certKey: '...my_cert_key...'
enabled: true
required: true
verify: true
version: tlsv1_2
threshold: 3.66
timeout: 5000
user: postgres
writeTimeout: 60000
logging:
payloads: false
maxRequestBodySize: 8.388608e+06
model:
nameHeader: true
proxy:
auth:
password: '...my_password...'
username: '...my_username...'
httpProxy:
host: '...my_host...'
port: 30633
httpsProxy:
host: '...my_host...'
port: 29606
noProxy: '...my_no_proxy...'
proxyScheme: http
responseStreaming: allow
route:
headers:
key:
fn::toJSON: value
hosts:
- foo.example.com
httpsRedirectStatusCode: 426
methods:
- '...'
model:
bodyParam: model
headerParam: x-model
pathParam: model_name
values:
- '...'
paths:
- '...'
preserveHost: false
protocols:
- '...'
regexPriority: 0
requestBuffering: true
responseBuffering: true
stripPath: true
tags:
- '...'
displayName: My GPT 5 model
enabled: true
formats:
- type: openai
labels:
key: value
managedBy:
key: value
name: my-gpt-5-model
policies:
- '...'
targets:
- allowAuthOverride: false
config:
databricks:
cacheReadCost: 8.45
cacheWriteCost: 3.62
cacheWriteCostLists:
- cost: 3.12
ttl: '...my_ttl...'
contextWindowFactors:
- above: '...my_above...'
inputFactor: 4.94
outputFactor: 9.43
embeddingsDimensions: 1.316728274e+09
inputCost: 9.06
maxTokens: 1.585442569e+09
outputCost: 7.78
serviceTierFactors:
- factor: 6.01
tier: '...my_tier...'
temperature: 3.33
topK: 8.96181225e+08
topP: 7.55
upstreamUrl: https://distant-antelope.com
workspaceInstanceId: '...my_workspace_instance_id...'
name: gpt-5-model
provider: azure-ai-se
semanticDescription: '...my_semantic_description...'
weight: 100
Example coming soon!
Create AiGatewayModel Resource
Resources are created with functions called constructors. To learn more about declaring and configuring resources, see Resources.
Constructor syntax
new AiGatewayModel(name: string, args: AiGatewayModelArgs, opts?: CustomResourceOptions);@overload
def AiGatewayModel(resource_name: str,
args: AiGatewayModelArgs,
opts: Optional[ResourceOptions] = None)
@overload
def AiGatewayModel(resource_name: str,
opts: Optional[ResourceOptions] = None,
gateway_id: Optional[str] = None,
api: Optional[AiGatewayModelApiArgs] = None,
model: Optional[AiGatewayModelModelArgs] = None)func NewAiGatewayModel(ctx *Context, name string, args AiGatewayModelArgs, opts ...ResourceOption) (*AiGatewayModel, error)public AiGatewayModel(string name, AiGatewayModelArgs args, CustomResourceOptions? opts = null)
public AiGatewayModel(String name, AiGatewayModelArgs args)
public AiGatewayModel(String name, AiGatewayModelArgs args, CustomResourceOptions options)
type: konnect:AiGatewayModel
properties: # The arguments to resource properties.
options: # Bag of options to control resource's behavior.
resource "konnect_ai_gateway_model" "name" {
# resource properties
}Parameters
- name string
- The unique name of the resource.
- args AiGatewayModelArgs
- The arguments to resource properties.
- opts CustomResourceOptions
- Bag of options to control resource's behavior.
- resource_name str
- The unique name of the resource.
- args AiGatewayModelArgs
- The arguments to resource properties.
- opts ResourceOptions
- Bag of options to control resource's behavior.
- ctx Context
- Context object for the current deployment.
- name string
- The unique name of the resource.
- args AiGatewayModelArgs
- The arguments to resource properties.
- opts ResourceOption
- Bag of options to control resource's behavior.
- name string
- The unique name of the resource.
- args AiGatewayModelArgs
- The arguments to resource properties.
- opts CustomResourceOptions
- Bag of options to control resource's behavior.
- name String
- The unique name of the resource.
- args AiGatewayModelArgs
- The arguments to resource properties.
- options CustomResourceOptions
- Bag of options to control resource's behavior.
AiGatewayModel Resource Properties
To learn more about resource properties and how to use them, see Inputs and Outputs in the Architecture and Concepts docs.
Inputs
In Python, inputs that are objects can be passed either as argument classes or as dictionary literals.
The AiGatewayModel resource accepts the following input properties:
- Gateway
Id string - The unique ID of the AI Gateway.
- Api
Ai
Gateway Model Api - Configuration for proxying asynchronous requests/responses to/from an AI Gateway model using the files and batches APIs.
- Model
Ai
Gateway Model Model - Configuration for proxying synchronous requests/responses to/from an AI Gateway model using generative APIs.
- Gateway
Id string - The unique ID of the AI Gateway.
- Api
Ai
Gateway Model Api Args - Configuration for proxying asynchronous requests/responses to/from an AI Gateway model using the files and batches APIs.
- Model
Ai
Gateway Model Model Args - Configuration for proxying synchronous requests/responses to/from an AI Gateway model using generative APIs.
- gateway_
id string - The unique ID of the AI Gateway.
- api object
- Configuration for proxying asynchronous requests/responses to/from an AI Gateway model using the files and batches APIs.
- model object
- Configuration for proxying synchronous requests/responses to/from an AI Gateway model using generative APIs.
- gateway
Id String - The unique ID of the AI Gateway.
- api
Ai
Gateway Model Api - Configuration for proxying asynchronous requests/responses to/from an AI Gateway model using the files and batches APIs.
- model
Ai
Gateway Model Model - Configuration for proxying synchronous requests/responses to/from an AI Gateway model using generative APIs.
- gateway
Id string - The unique ID of the AI Gateway.
- api
Ai
Gateway Model Api - Configuration for proxying asynchronous requests/responses to/from an AI Gateway model using the files and batches APIs.
- model
Ai
Gateway Model Model - Configuration for proxying synchronous requests/responses to/from an AI Gateway model using generative APIs.
- gateway_
id str - The unique ID of the AI Gateway.
- api
Ai
Gateway Model Api Args - Configuration for proxying asynchronous requests/responses to/from an AI Gateway model using the files and batches APIs.
- model
Ai
Gateway Model Model Args - Configuration for proxying synchronous requests/responses to/from an AI Gateway model using generative APIs.
- gateway
Id String - The unique ID of the AI Gateway.
- api Property Map
- Configuration for proxying asynchronous requests/responses to/from an AI Gateway model using the files and batches APIs.
- model Property Map
- Configuration for proxying synchronous requests/responses to/from an AI Gateway model using generative APIs.
Outputs
All input properties are implicitly available as output properties. Additionally, the AiGatewayModel resource produces the following output properties:
- Created
At string - An ISO-8601 timestamp representation of entity creation date.
- Display
Name string - The display name for this model instance.
- Enabled bool
- Whether the model is enabled.
- Id string
- The provider-assigned unique ID for this managed resource.
- Name string
- A user-defined unique identifier for this model, used as a stable human-readable reference. This value is immutable after creation.
- Updated
At string - An ISO-8601 timestamp representation of entity update date.
- Created
At string - An ISO-8601 timestamp representation of entity creation date.
- Display
Name string - The display name for this model instance.
- Enabled bool
- Whether the model is enabled.
- Id string
- The provider-assigned unique ID for this managed resource.
- Name string
- A user-defined unique identifier for this model, used as a stable human-readable reference. This value is immutable after creation.
- Updated
At string - An ISO-8601 timestamp representation of entity update date.
- created_
at string - An ISO-8601 timestamp representation of entity creation date.
- display_
name string - The display name for this model instance.
- enabled bool
- Whether the model is enabled.
- id string
- The provider-assigned unique ID for this managed resource.
- name string
- A user-defined unique identifier for this model, used as a stable human-readable reference. This value is immutable after creation.
- updated_
at string - An ISO-8601 timestamp representation of entity update date.
- created
At String - An ISO-8601 timestamp representation of entity creation date.
- display
Name String - The display name for this model instance.
- enabled Boolean
- Whether the model is enabled.
- id String
- The provider-assigned unique ID for this managed resource.
- name String
- A user-defined unique identifier for this model, used as a stable human-readable reference. This value is immutable after creation.
- updated
At String - An ISO-8601 timestamp representation of entity update date.
- created
At string - An ISO-8601 timestamp representation of entity creation date.
- display
Name string - The display name for this model instance.
- enabled boolean
- Whether the model is enabled.
- id string
- The provider-assigned unique ID for this managed resource.
- name string
- A user-defined unique identifier for this model, used as a stable human-readable reference. This value is immutable after creation.
- updated
At string - An ISO-8601 timestamp representation of entity update date.
- created_
at str - An ISO-8601 timestamp representation of entity creation date.
- display_
name str - The display name for this model instance.
- enabled bool
- Whether the model is enabled.
- id str
- The provider-assigned unique ID for this managed resource.
- name str
- A user-defined unique identifier for this model, used as a stable human-readable reference. This value is immutable after creation.
- updated_
at str - An ISO-8601 timestamp representation of entity update date.
- created
At String - An ISO-8601 timestamp representation of entity creation date.
- display
Name String - The display name for this model instance.
- enabled Boolean
- Whether the model is enabled.
- id String
- The provider-assigned unique ID for this managed resource.
- name String
- A user-defined unique identifier for this model, used as a stable human-readable reference. This value is immutable after creation.
- updated
At String - An ISO-8601 timestamp representation of entity update date.
Look up Existing AiGatewayModel Resource
Get an existing AiGatewayModel resource’s state with the given name, ID, and optional extra properties used to qualify the lookup.
public static get(name: string, id: Input<ID>, state?: AiGatewayModelState, opts?: CustomResourceOptions): AiGatewayModel@staticmethod
def get(resource_name: str,
id: str,
opts: Optional[ResourceOptions] = None,
api: Optional[AiGatewayModelApiArgs] = None,
created_at: Optional[str] = None,
display_name: Optional[str] = None,
enabled: Optional[bool] = None,
gateway_id: Optional[str] = None,
model: Optional[AiGatewayModelModelArgs] = None,
name: Optional[str] = None,
updated_at: Optional[str] = None) -> AiGatewayModelfunc GetAiGatewayModel(ctx *Context, name string, id IDInput, state *AiGatewayModelState, opts ...ResourceOption) (*AiGatewayModel, error)public static AiGatewayModel Get(string name, Input<string> id, AiGatewayModelState? state, CustomResourceOptions? opts = null)public static AiGatewayModel get(String name, Output<String> id, AiGatewayModelState state, CustomResourceOptions options)resources: _: type: konnect:AiGatewayModel get: id: ${id}import {
to = konnect_ai_gateway_model.example
id = "${id}"
}
- name
- The unique name of the resulting resource.
- id
- The unique provider ID of the resource to lookup.
- state
- Any extra arguments used during the lookup.
- opts
- A bag of options that control this resource's behavior.
- resource_name
- The unique name of the resulting resource.
- id
- The unique provider ID of the resource to lookup.
- name
- The unique name of the resulting resource.
- id
- The unique provider ID of the resource to lookup.
- state
- Any extra arguments used during the lookup.
- opts
- A bag of options that control this resource's behavior.
- name
- The unique name of the resulting resource.
- id
- The unique provider ID of the resource to lookup.
- state
- Any extra arguments used during the lookup.
- opts
- A bag of options that control this resource's behavior.
- name
- The unique name of the resulting resource.
- id
- The unique provider ID of the resource to lookup.
- state
- Any extra arguments used during the lookup.
- opts
- A bag of options that control this resource's behavior.
- Api
Ai
Gateway Model Api - Configuration for proxying asynchronous requests/responses to/from an AI Gateway model using the files and batches APIs.
- Created
At string - An ISO-8601 timestamp representation of entity creation date.
- Display
Name string - The display name for this model instance.
- Enabled bool
- Whether the model is enabled.
- Gateway
Id string - The unique ID of the AI Gateway.
- Model
Ai
Gateway Model Model - Configuration for proxying synchronous requests/responses to/from an AI Gateway model using generative APIs.
- Name string
- A user-defined unique identifier for this model, used as a stable human-readable reference. This value is immutable after creation.
- Updated
At string - An ISO-8601 timestamp representation of entity update date.
- Api
Ai
Gateway Model Api Args - Configuration for proxying asynchronous requests/responses to/from an AI Gateway model using the files and batches APIs.
- Created
At string - An ISO-8601 timestamp representation of entity creation date.
- Display
Name string - The display name for this model instance.
- Enabled bool
- Whether the model is enabled.
- Gateway
Id string - The unique ID of the AI Gateway.
- Model
Ai
Gateway Model Model Args - Configuration for proxying synchronous requests/responses to/from an AI Gateway model using generative APIs.
- Name string
- A user-defined unique identifier for this model, used as a stable human-readable reference. This value is immutable after creation.
- Updated
At string - An ISO-8601 timestamp representation of entity update date.
- api object
- Configuration for proxying asynchronous requests/responses to/from an AI Gateway model using the files and batches APIs.
- created_
at string - An ISO-8601 timestamp representation of entity creation date.
- display_
name string - The display name for this model instance.
- enabled bool
- Whether the model is enabled.
- gateway_
id string - The unique ID of the AI Gateway.
- model object
- Configuration for proxying synchronous requests/responses to/from an AI Gateway model using generative APIs.
- name string
- A user-defined unique identifier for this model, used as a stable human-readable reference. This value is immutable after creation.
- updated_
at string - An ISO-8601 timestamp representation of entity update date.
- api
Ai
Gateway Model Api - Configuration for proxying asynchronous requests/responses to/from an AI Gateway model using the files and batches APIs.
- created
At String - An ISO-8601 timestamp representation of entity creation date.
- display
Name String - The display name for this model instance.
- enabled Boolean
- Whether the model is enabled.
- gateway
Id String - The unique ID of the AI Gateway.
- model
Ai
Gateway Model Model - Configuration for proxying synchronous requests/responses to/from an AI Gateway model using generative APIs.
- name String
- A user-defined unique identifier for this model, used as a stable human-readable reference. This value is immutable after creation.
- updated
At String - An ISO-8601 timestamp representation of entity update date.
- api
Ai
Gateway Model Api - Configuration for proxying asynchronous requests/responses to/from an AI Gateway model using the files and batches APIs.
- created
At string - An ISO-8601 timestamp representation of entity creation date.
- display
Name string - The display name for this model instance.
- enabled boolean
- Whether the model is enabled.
- gateway
Id string - The unique ID of the AI Gateway.
- model
Ai
Gateway Model Model - Configuration for proxying synchronous requests/responses to/from an AI Gateway model using generative APIs.
- name string
- A user-defined unique identifier for this model, used as a stable human-readable reference. This value is immutable after creation.
- updated
At string - An ISO-8601 timestamp representation of entity update date.
- api
Ai
Gateway Model Api Args - Configuration for proxying asynchronous requests/responses to/from an AI Gateway model using the files and batches APIs.
- created_
at str - An ISO-8601 timestamp representation of entity creation date.
- display_
name str - The display name for this model instance.
- enabled bool
- Whether the model is enabled.
- gateway_
id str - The unique ID of the AI Gateway.
- model
Ai
Gateway Model Model Args - Configuration for proxying synchronous requests/responses to/from an AI Gateway model using generative APIs.
- name str
- A user-defined unique identifier for this model, used as a stable human-readable reference. This value is immutable after creation.
- updated_
at str - An ISO-8601 timestamp representation of entity update date.
- api Property Map
- Configuration for proxying asynchronous requests/responses to/from an AI Gateway model using the files and batches APIs.
- created
At String - An ISO-8601 timestamp representation of entity creation date.
- display
Name String - The display name for this model instance.
- enabled Boolean
- Whether the model is enabled.
- gateway
Id String - The unique ID of the AI Gateway.
- model Property Map
- Configuration for proxying synchronous requests/responses to/from an AI Gateway model using generative APIs.
- name String
- A user-defined unique identifier for this model, used as a stable human-readable reference. This value is immutable after creation.
- updated
At String - An ISO-8601 timestamp representation of entity update date.
Supporting Types
AiGatewayModelApi, AiGatewayModelApiArgs
- Access
Ai
Gateway Model Api Access - Access control configuration for a model.
- Capabilities List<string>
- List of AI capabilities enabled for this API model. Not Null
- Config
Ai
Gateway Model Api Config - Routing, logging, and load balancing configuration for the model. Not Null
- Created
At string - An ISO-8601 timestamp representation of entity creation date.
- Display
Name string - The display name for this model instance. Not Null
- Enabled bool
- Whether the model is enabled. Default: true
- Formats
List<Ai
Gateway Model Api Format> - List of request/response formats supported by this model. Not Null
- Id string
- Contains a unique identifier used for this resource.
- Labels Dictionary<string, string>
- Public labels store information about an entity that can be used for filtering a list of objects.
- Managed
By Dictionary<string, string> - Name string
- A user-defined unique identifier for this model, used as a stable human-readable reference. This value is immutable after creation.
- Policies List<string>
- List of policy references.
- Targets
List<Ai
Gateway Model Api Target> - One or more backend models that this model entry routes to. Not Null
- Updated
At string - An ISO-8601 timestamp representation of entity update date.
- Access
Ai
Gateway Model Api Access - Access control configuration for a model.
- Capabilities []string
- List of AI capabilities enabled for this API model. Not Null
- Config
Ai
Gateway Model Api Config - Routing, logging, and load balancing configuration for the model. Not Null
- Created
At string - An ISO-8601 timestamp representation of entity creation date.
- Display
Name string - The display name for this model instance. Not Null
- Enabled bool
- Whether the model is enabled. Default: true
- Formats
[]Ai
Gateway Model Api Format - List of request/response formats supported by this model. Not Null
- Id string
- Contains a unique identifier used for this resource.
- Labels map[string]string
- Public labels store information about an entity that can be used for filtering a list of objects.
- Managed
By map[string]string - Name string
- A user-defined unique identifier for this model, used as a stable human-readable reference. This value is immutable after creation.
- Policies []string
- List of policy references.
- Targets
[]Ai
Gateway Model Api Target - One or more backend models that this model entry routes to. Not Null
- Updated
At string - An ISO-8601 timestamp representation of entity update date.
- access object
- Access control configuration for a model.
- capabilities list(string)
- List of AI capabilities enabled for this API model. Not Null
- config object
- Routing, logging, and load balancing configuration for the model. Not Null
- created_
at string - An ISO-8601 timestamp representation of entity creation date.
- display_
name string - The display name for this model instance. Not Null
- enabled bool
- Whether the model is enabled. Default: true
- formats list(object)
- List of request/response formats supported by this model. Not Null
- id string
- Contains a unique identifier used for this resource.
- labels map(string)
- Public labels store information about an entity that can be used for filtering a list of objects.
- managed_
by map(string) - name string
- A user-defined unique identifier for this model, used as a stable human-readable reference. This value is immutable after creation.
- policies list(string)
- List of policy references.
- targets list(object)
- One or more backend models that this model entry routes to. Not Null
- updated_
at string - An ISO-8601 timestamp representation of entity update date.
- access
Ai
Gateway Model Api Access - Access control configuration for a model.
- capabilities List<String>
- List of AI capabilities enabled for this API model. Not Null
- config
Ai
Gateway Model Api Config - Routing, logging, and load balancing configuration for the model. Not Null
- created
At String - An ISO-8601 timestamp representation of entity creation date.
- display
Name String - The display name for this model instance. Not Null
- enabled Boolean
- Whether the model is enabled. Default: true
- formats
List<Ai
Gateway Model Api Format> - List of request/response formats supported by this model. Not Null
- id String
- Contains a unique identifier used for this resource.
- labels Map<String,String>
- Public labels store information about an entity that can be used for filtering a list of objects.
- managed
By Map<String,String> - name String
- A user-defined unique identifier for this model, used as a stable human-readable reference. This value is immutable after creation.
- policies List<String>
- List of policy references.
- targets
List<Ai
Gateway Model Api Target> - One or more backend models that this model entry routes to. Not Null
- updated
At String - An ISO-8601 timestamp representation of entity update date.
- access
Ai
Gateway Model Api Access - Access control configuration for a model.
- capabilities string[]
- List of AI capabilities enabled for this API model. Not Null
- config
Ai
Gateway Model Api Config - Routing, logging, and load balancing configuration for the model. Not Null
- created
At string - An ISO-8601 timestamp representation of entity creation date.
- display
Name string - The display name for this model instance. Not Null
- enabled boolean
- Whether the model is enabled. Default: true
- formats
Ai
Gateway Model Api Format[] - List of request/response formats supported by this model. Not Null
- id string
- Contains a unique identifier used for this resource.
- labels {[key: string]: string}
- Public labels store information about an entity that can be used for filtering a list of objects.
- managed
By {[key: string]: string} - name string
- A user-defined unique identifier for this model, used as a stable human-readable reference. This value is immutable after creation.
- policies string[]
- List of policy references.
- targets
Ai
Gateway Model Api Target[] - One or more backend models that this model entry routes to. Not Null
- updated
At string - An ISO-8601 timestamp representation of entity update date.
- access
Ai
Gateway Model Api Access - Access control configuration for a model.
- capabilities Sequence[str]
- List of AI capabilities enabled for this API model. Not Null
- config
Ai
Gateway Model Api Config - Routing, logging, and load balancing configuration for the model. Not Null
- created_
at str - An ISO-8601 timestamp representation of entity creation date.
- display_
name str - The display name for this model instance. Not Null
- enabled bool
- Whether the model is enabled. Default: true
- formats
Sequence[Ai
Gateway Model Api Format] - List of request/response formats supported by this model. Not Null
- id str
- Contains a unique identifier used for this resource.
- labels Mapping[str, str]
- Public labels store information about an entity that can be used for filtering a list of objects.
- managed_
by Mapping[str, str] - name str
- A user-defined unique identifier for this model, used as a stable human-readable reference. This value is immutable after creation.
- policies Sequence[str]
- List of policy references.
- targets
Sequence[Ai
Gateway Model Api Target] - One or more backend models that this model entry routes to. Not Null
- updated_
at str - An ISO-8601 timestamp representation of entity update date.
- access Property Map
- Access control configuration for a model.
- capabilities List<String>
- List of AI capabilities enabled for this API model. Not Null
- config Property Map
- Routing, logging, and load balancing configuration for the model. Not Null
- created
At String - An ISO-8601 timestamp representation of entity creation date.
- display
Name String - The display name for this model instance. Not Null
- enabled Boolean
- Whether the model is enabled. Default: true
- formats List<Property Map>
- List of request/response formats supported by this model. Not Null
- id String
- Contains a unique identifier used for this resource.
- labels Map<String>
- Public labels store information about an entity that can be used for filtering a list of objects.
- managed
By Map<String> - name String
- A user-defined unique identifier for this model, used as a stable human-readable reference. This value is immutable after creation.
- policies List<String>
- List of policy references.
- targets List<Property Map>
- One or more backend models that this model entry routes to. Not Null
- updated
At String - An ISO-8601 timestamp representation of entity update date.
AiGatewayModelApiAccess, AiGatewayModelApiAccessArgs
- Acls
Ai
Gateway Model Api Access Acls - Access control rules. Configure exactly one of
allowordeny. - Auth
Strategies List<string> - List of auth strategies for granting access to the model. At most 1 auth strategy of each auth strategy type can be referenced.
- Identity
Providers List<string> - List of identity providers for granting access to the model. At most 1 identity provider of each identity provider type can be referenced.
- Acls
Ai
Gateway Model Api Access Acls - Access control rules. Configure exactly one of
allowordeny. - Auth
Strategies []string - List of auth strategies for granting access to the model. At most 1 auth strategy of each auth strategy type can be referenced.
- Identity
Providers []string - List of identity providers for granting access to the model. At most 1 identity provider of each identity provider type can be referenced.
- acls object
- Access control rules. Configure exactly one of
allowordeny. - auth_
strategies list(string) - List of auth strategies for granting access to the model. At most 1 auth strategy of each auth strategy type can be referenced.
- identity_
providers list(string) - List of identity providers for granting access to the model. At most 1 identity provider of each identity provider type can be referenced.
- acls
Ai
Gateway Model Api Access Acls - Access control rules. Configure exactly one of
allowordeny. - auth
Strategies List<String> - List of auth strategies for granting access to the model. At most 1 auth strategy of each auth strategy type can be referenced.
- identity
Providers List<String> - List of identity providers for granting access to the model. At most 1 identity provider of each identity provider type can be referenced.
- acls
Ai
Gateway Model Api Access Acls - Access control rules. Configure exactly one of
allowordeny. - auth
Strategies string[] - List of auth strategies for granting access to the model. At most 1 auth strategy of each auth strategy type can be referenced.
- identity
Providers string[] - List of identity providers for granting access to the model. At most 1 identity provider of each identity provider type can be referenced.
- acls
Ai
Gateway Model Api Access Acls - Access control rules. Configure exactly one of
allowordeny. - auth_
strategies Sequence[str] - List of auth strategies for granting access to the model. At most 1 auth strategy of each auth strategy type can be referenced.
- identity_
providers Sequence[str] - List of identity providers for granting access to the model. At most 1 identity provider of each identity provider type can be referenced.
- acls Property Map
- Access control rules. Configure exactly one of
allowordeny. - auth
Strategies List<String> - List of auth strategies for granting access to the model. At most 1 auth strategy of each auth strategy type can be referenced.
- identity
Providers List<String> - List of identity providers for granting access to the model. At most 1 identity provider of each identity provider type can be referenced.
AiGatewayModelApiAccessAcls, AiGatewayModelApiAccessAclsArgs
AiGatewayModelApiConfig, AiGatewayModelApiConfigArgs
- Balancer
Ai
Gateway Model Api Config Balancer - Configuration for a model's load balancer when multiple target models are configured.
- Logging
Ai
Gateway Model Api Config Logging - Configuration for AI Gateway logging.
- Max
Request doubleBody Size - Maximum size of request body to parse. Set to 0 for unlimited. Default: 8388608
- Proxy
Ai
Gateway Model Api Config Proxy - HTTP/HTTPS proxy configuration for outbound requests to the upstream AI provider.
- Response
Streaming string - possible known values include one of ["allow", "always", "deny"]; Default: "allow"
- Route
Ai
Gateway Model Api Config Route - Configuration for an AI Gateway route. Not Null
- Balancer
Ai
Gateway Model Api Config Balancer - Configuration for a model's load balancer when multiple target models are configured.
- Logging
Ai
Gateway Model Api Config Logging - Configuration for AI Gateway logging.
- Max
Request float64Body Size - Maximum size of request body to parse. Set to 0 for unlimited. Default: 8388608
- Proxy
Ai
Gateway Model Api Config Proxy - HTTP/HTTPS proxy configuration for outbound requests to the upstream AI provider.
- Response
Streaming string - possible known values include one of ["allow", "always", "deny"]; Default: "allow"
- Route
Ai
Gateway Model Api Config Route - Configuration for an AI Gateway route. Not Null
- balancer object
- Configuration for a model's load balancer when multiple target models are configured.
- logging object
- Configuration for AI Gateway logging.
- max_
request_ numberbody_ size - Maximum size of request body to parse. Set to 0 for unlimited. Default: 8388608
- proxy object
- HTTP/HTTPS proxy configuration for outbound requests to the upstream AI provider.
- response_
streaming string - possible known values include one of ["allow", "always", "deny"]; Default: "allow"
- route object
- Configuration for an AI Gateway route. Not Null
- balancer
Ai
Gateway Model Api Config Balancer - Configuration for a model's load balancer when multiple target models are configured.
- logging
Ai
Gateway Model Api Config Logging - Configuration for AI Gateway logging.
- max
Request DoubleBody Size - Maximum size of request body to parse. Set to 0 for unlimited. Default: 8388608
- proxy
Ai
Gateway Model Api Config Proxy - HTTP/HTTPS proxy configuration for outbound requests to the upstream AI provider.
- response
Streaming String - possible known values include one of ["allow", "always", "deny"]; Default: "allow"
- route
Ai
Gateway Model Api Config Route - Configuration for an AI Gateway route. Not Null
- balancer
Ai
Gateway Model Api Config Balancer - Configuration for a model's load balancer when multiple target models are configured.
- logging
Ai
Gateway Model Api Config Logging - Configuration for AI Gateway logging.
- max
Request numberBody Size - Maximum size of request body to parse. Set to 0 for unlimited. Default: 8388608
- proxy
Ai
Gateway Model Api Config Proxy - HTTP/HTTPS proxy configuration for outbound requests to the upstream AI provider.
- response
Streaming string - possible known values include one of ["allow", "always", "deny"]; Default: "allow"
- route
Ai
Gateway Model Api Config Route - Configuration for an AI Gateway route. Not Null
- balancer
Ai
Gateway Model Api Config Balancer - Configuration for a model's load balancer when multiple target models are configured.
- logging
Ai
Gateway Model Api Config Logging - Configuration for AI Gateway logging.
- max_
request_ floatbody_ size - Maximum size of request body to parse. Set to 0 for unlimited. Default: 8388608
- proxy
Ai
Gateway Model Api Config Proxy - HTTP/HTTPS proxy configuration for outbound requests to the upstream AI provider.
- response_
streaming str - possible known values include one of ["allow", "always", "deny"]; Default: "allow"
- route
Ai
Gateway Model Api Config Route - Configuration for an AI Gateway route. Not Null
- balancer Property Map
- Configuration for a model's load balancer when multiple target models are configured.
- logging Property Map
- Configuration for AI Gateway logging.
- max
Request NumberBody Size - Maximum size of request body to parse. Set to 0 for unlimited. Default: 8388608
- proxy Property Map
- HTTP/HTTPS proxy configuration for outbound requests to the upstream AI provider.
- response
Streaming String - possible known values include one of ["allow", "always", "deny"]; Default: "allow"
- route Property Map
- Configuration for an AI Gateway route. Not Null
AiGatewayModelApiConfigBalancer, AiGatewayModelApiConfigBalancerArgs
- Consistent
Hashing AiGateway Model Api Config Balancer Consistent Hashing - Least
Connections AiGateway Model Api Config Balancer Least Connections - Lowest
Latency AiGateway Model Api Config Balancer Lowest Latency - Lowest
Usage AiGateway Model Api Config Balancer Lowest Usage - Priority
Ai
Gateway Model Api Config Balancer Priority - Round
Robin AiGateway Model Api Config Balancer Round Robin - Semantic
Ai
Gateway Model Api Config Balancer Semantic
- Consistent
Hashing AiGateway Model Api Config Balancer Consistent Hashing - Least
Connections AiGateway Model Api Config Balancer Least Connections - Lowest
Latency AiGateway Model Api Config Balancer Lowest Latency - Lowest
Usage AiGateway Model Api Config Balancer Lowest Usage - Priority
Ai
Gateway Model Api Config Balancer Priority - Round
Robin AiGateway Model Api Config Balancer Round Robin - Semantic
Ai
Gateway Model Api Config Balancer Semantic
- consistent
Hashing AiGateway Model Api Config Balancer Consistent Hashing - least
Connections AiGateway Model Api Config Balancer Least Connections - lowest
Latency AiGateway Model Api Config Balancer Lowest Latency - lowest
Usage AiGateway Model Api Config Balancer Lowest Usage - priority
Ai
Gateway Model Api Config Balancer Priority - round
Robin AiGateway Model Api Config Balancer Round Robin - semantic
Ai
Gateway Model Api Config Balancer Semantic
- consistent
Hashing AiGateway Model Api Config Balancer Consistent Hashing - least
Connections AiGateway Model Api Config Balancer Least Connections - lowest
Latency AiGateway Model Api Config Balancer Lowest Latency - lowest
Usage AiGateway Model Api Config Balancer Lowest Usage - priority
Ai
Gateway Model Api Config Balancer Priority - round
Robin AiGateway Model Api Config Balancer Round Robin - semantic
Ai
Gateway Model Api Config Balancer Semantic
- consistent_
hashing AiGateway Model Api Config Balancer Consistent Hashing - least_
connections AiGateway Model Api Config Balancer Least Connections - lowest_
latency AiGateway Model Api Config Balancer Lowest Latency - lowest_
usage AiGateway Model Api Config Balancer Lowest Usage - priority
Ai
Gateway Model Api Config Balancer Priority - round_
robin AiGateway Model Api Config Balancer Round Robin - semantic
Ai
Gateway Model Api Config Balancer Semantic
AiGatewayModelApiConfigBalancerConsistentHashing, AiGatewayModelApiConfigBalancerConsistentHashingArgs
- Connect
Timeout double - Default: 60000
- Fail
Timeout double - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - Failover
Criterias List<string> - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- Hash
On stringHeader - The header to use for consistent-hashing. Default: "X-Kong-LLM-Request-ID"
- Max
Fails double - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - Read
Timeout double - Default: 60000
- Retries double
- The number of retries to execute upon failure to proxy. Default: 5
- Slots double
- The number of slots in the load balancer algorithm. Default: 10000
- Write
Timeout double - Default: 60000
- Connect
Timeout float64 - Default: 60000
- Fail
Timeout float64 - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - Failover
Criterias []string - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- Hash
On stringHeader - The header to use for consistent-hashing. Default: "X-Kong-LLM-Request-ID"
- Max
Fails float64 - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - Read
Timeout float64 - Default: 60000
- Retries float64
- The number of retries to execute upon failure to proxy. Default: 5
- Slots float64
- The number of slots in the load balancer algorithm. Default: 10000
- Write
Timeout float64 - Default: 60000
- connect_
timeout number - Default: 60000
- fail_
timeout number - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover_
criterias list(string) - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- hash_
on_ stringheader - The header to use for consistent-hashing. Default: "X-Kong-LLM-Request-ID"
- max_
fails number - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read_
timeout number - Default: 60000
- retries number
- The number of retries to execute upon failure to proxy. Default: 5
- slots number
- The number of slots in the load balancer algorithm. Default: 10000
- write_
timeout number - Default: 60000
- connect
Timeout Double - Default: 60000
- fail
Timeout Double - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover
Criterias List<String> - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- hash
On StringHeader - The header to use for consistent-hashing. Default: "X-Kong-LLM-Request-ID"
- max
Fails Double - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read
Timeout Double - Default: 60000
- retries Double
- The number of retries to execute upon failure to proxy. Default: 5
- slots Double
- The number of slots in the load balancer algorithm. Default: 10000
- write
Timeout Double - Default: 60000
- connect
Timeout number - Default: 60000
- fail
Timeout number - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover
Criterias string[] - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- hash
On stringHeader - The header to use for consistent-hashing. Default: "X-Kong-LLM-Request-ID"
- max
Fails number - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read
Timeout number - Default: 60000
- retries number
- The number of retries to execute upon failure to proxy. Default: 5
- slots number
- The number of slots in the load balancer algorithm. Default: 10000
- write
Timeout number - Default: 60000
- connect_
timeout float - Default: 60000
- fail_
timeout float - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover_
criterias Sequence[str] - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- hash_
on_ strheader - The header to use for consistent-hashing. Default: "X-Kong-LLM-Request-ID"
- max_
fails float - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read_
timeout float - Default: 60000
- retries float
- The number of retries to execute upon failure to proxy. Default: 5
- slots float
- The number of slots in the load balancer algorithm. Default: 10000
- write_
timeout float - Default: 60000
- connect
Timeout Number - Default: 60000
- fail
Timeout Number - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover
Criterias List<String> - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- hash
On StringHeader - The header to use for consistent-hashing. Default: "X-Kong-LLM-Request-ID"
- max
Fails Number - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read
Timeout Number - Default: 60000
- retries Number
- The number of retries to execute upon failure to proxy. Default: 5
- slots Number
- The number of slots in the load balancer algorithm. Default: 10000
- write
Timeout Number - Default: 60000
AiGatewayModelApiConfigBalancerLeastConnections, AiGatewayModelApiConfigBalancerLeastConnectionsArgs
- Connect
Timeout double - Default: 60000
- Fail
Timeout double - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - Failover
Criterias List<string> - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- Max
Fails double - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - Read
Timeout double - Default: 60000
- Retries double
- The number of retries to execute upon failure to proxy. Default: 5
- Slots double
- The number of slots in the load balancer algorithm. Default: 10000
- Write
Timeout double - Default: 60000
- Connect
Timeout float64 - Default: 60000
- Fail
Timeout float64 - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - Failover
Criterias []string - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- Max
Fails float64 - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - Read
Timeout float64 - Default: 60000
- Retries float64
- The number of retries to execute upon failure to proxy. Default: 5
- Slots float64
- The number of slots in the load balancer algorithm. Default: 10000
- Write
Timeout float64 - Default: 60000
- connect_
timeout number - Default: 60000
- fail_
timeout number - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover_
criterias list(string) - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- max_
fails number - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read_
timeout number - Default: 60000
- retries number
- The number of retries to execute upon failure to proxy. Default: 5
- slots number
- The number of slots in the load balancer algorithm. Default: 10000
- write_
timeout number - Default: 60000
- connect
Timeout Double - Default: 60000
- fail
Timeout Double - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover
Criterias List<String> - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- max
Fails Double - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read
Timeout Double - Default: 60000
- retries Double
- The number of retries to execute upon failure to proxy. Default: 5
- slots Double
- The number of slots in the load balancer algorithm. Default: 10000
- write
Timeout Double - Default: 60000
- connect
Timeout number - Default: 60000
- fail
Timeout number - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover
Criterias string[] - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- max
Fails number - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read
Timeout number - Default: 60000
- retries number
- The number of retries to execute upon failure to proxy. Default: 5
- slots number
- The number of slots in the load balancer algorithm. Default: 10000
- write
Timeout number - Default: 60000
- connect_
timeout float - Default: 60000
- fail_
timeout float - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover_
criterias Sequence[str] - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- max_
fails float - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read_
timeout float - Default: 60000
- retries float
- The number of retries to execute upon failure to proxy. Default: 5
- slots float
- The number of slots in the load balancer algorithm. Default: 10000
- write_
timeout float - Default: 60000
- connect
Timeout Number - Default: 60000
- fail
Timeout Number - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover
Criterias List<String> - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- max
Fails Number - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read
Timeout Number - Default: 60000
- retries Number
- The number of retries to execute upon failure to proxy. Default: 5
- slots Number
- The number of slots in the load balancer algorithm. Default: 10000
- write
Timeout Number - Default: 60000
AiGatewayModelApiConfigBalancerLowestLatency, AiGatewayModelApiConfigBalancerLowestLatencyArgs
- Connect
Timeout double - Default: 60000
- Fail
Timeout double - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - Failover
Criterias List<string> - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- Latency
Strategy string - What metrics to use for latency. Available values are:
tpot(time-per-output-token) ande2e. possible known values include one of ["e2e", "tpot"]; Default: "tpot" - Max
Fails double - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - Read
Timeout double - Default: 60000
- Retries double
- The number of retries to execute upon failure to proxy. Default: 5
- Slots double
- The number of slots in the load balancer algorithm. Default: 10000
- Write
Timeout double - Default: 60000
- Connect
Timeout float64 - Default: 60000
- Fail
Timeout float64 - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - Failover
Criterias []string - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- Latency
Strategy string - What metrics to use for latency. Available values are:
tpot(time-per-output-token) ande2e. possible known values include one of ["e2e", "tpot"]; Default: "tpot" - Max
Fails float64 - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - Read
Timeout float64 - Default: 60000
- Retries float64
- The number of retries to execute upon failure to proxy. Default: 5
- Slots float64
- The number of slots in the load balancer algorithm. Default: 10000
- Write
Timeout float64 - Default: 60000
- connect_
timeout number - Default: 60000
- fail_
timeout number - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover_
criterias list(string) - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- latency_
strategy string - What metrics to use for latency. Available values are:
tpot(time-per-output-token) ande2e. possible known values include one of ["e2e", "tpot"]; Default: "tpot" - max_
fails number - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read_
timeout number - Default: 60000
- retries number
- The number of retries to execute upon failure to proxy. Default: 5
- slots number
- The number of slots in the load balancer algorithm. Default: 10000
- write_
timeout number - Default: 60000
- connect
Timeout Double - Default: 60000
- fail
Timeout Double - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover
Criterias List<String> - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- latency
Strategy String - What metrics to use for latency. Available values are:
tpot(time-per-output-token) ande2e. possible known values include one of ["e2e", "tpot"]; Default: "tpot" - max
Fails Double - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read
Timeout Double - Default: 60000
- retries Double
- The number of retries to execute upon failure to proxy. Default: 5
- slots Double
- The number of slots in the load balancer algorithm. Default: 10000
- write
Timeout Double - Default: 60000
- connect
Timeout number - Default: 60000
- fail
Timeout number - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover
Criterias string[] - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- latency
Strategy string - What metrics to use for latency. Available values are:
tpot(time-per-output-token) ande2e. possible known values include one of ["e2e", "tpot"]; Default: "tpot" - max
Fails number - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read
Timeout number - Default: 60000
- retries number
- The number of retries to execute upon failure to proxy. Default: 5
- slots number
- The number of slots in the load balancer algorithm. Default: 10000
- write
Timeout number - Default: 60000
- connect_
timeout float - Default: 60000
- fail_
timeout float - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover_
criterias Sequence[str] - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- latency_
strategy str - What metrics to use for latency. Available values are:
tpot(time-per-output-token) ande2e. possible known values include one of ["e2e", "tpot"]; Default: "tpot" - max_
fails float - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read_
timeout float - Default: 60000
- retries float
- The number of retries to execute upon failure to proxy. Default: 5
- slots float
- The number of slots in the load balancer algorithm. Default: 10000
- write_
timeout float - Default: 60000
- connect
Timeout Number - Default: 60000
- fail
Timeout Number - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover
Criterias List<String> - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- latency
Strategy String - What metrics to use for latency. Available values are:
tpot(time-per-output-token) ande2e. possible known values include one of ["e2e", "tpot"]; Default: "tpot" - max
Fails Number - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read
Timeout Number - Default: 60000
- retries Number
- The number of retries to execute upon failure to proxy. Default: 5
- slots Number
- The number of slots in the load balancer algorithm. Default: 10000
- write
Timeout Number - Default: 60000
AiGatewayModelApiConfigBalancerLowestUsage, AiGatewayModelApiConfigBalancerLowestUsageArgs
- Connect
Timeout double - Default: 60000
- Fail
Timeout double - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - Failover
Criterias List<string> - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- Max
Fails double - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - Read
Timeout double - Default: 60000
- Retries double
- The number of retries to execute upon failure to proxy. Default: 5
- Slots double
- The number of slots in the load balancer algorithm. Default: 10000
- Tokens
Count stringStrategy - Methodology to use for token usage calculation. possible known values include one of ["completion-tokens", "cost", "llm-accuracy", "prompt-tokens", "total-tokens"]; Default: "total-tokens"
- Write
Timeout double - Default: 60000
- Connect
Timeout float64 - Default: 60000
- Fail
Timeout float64 - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - Failover
Criterias []string - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- Max
Fails float64 - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - Read
Timeout float64 - Default: 60000
- Retries float64
- The number of retries to execute upon failure to proxy. Default: 5
- Slots float64
- The number of slots in the load balancer algorithm. Default: 10000
- Tokens
Count stringStrategy - Methodology to use for token usage calculation. possible known values include one of ["completion-tokens", "cost", "llm-accuracy", "prompt-tokens", "total-tokens"]; Default: "total-tokens"
- Write
Timeout float64 - Default: 60000
- connect_
timeout number - Default: 60000
- fail_
timeout number - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover_
criterias list(string) - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- max_
fails number - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read_
timeout number - Default: 60000
- retries number
- The number of retries to execute upon failure to proxy. Default: 5
- slots number
- The number of slots in the load balancer algorithm. Default: 10000
- tokens_
count_ stringstrategy - Methodology to use for token usage calculation. possible known values include one of ["completion-tokens", "cost", "llm-accuracy", "prompt-tokens", "total-tokens"]; Default: "total-tokens"
- write_
timeout number - Default: 60000
- connect
Timeout Double - Default: 60000
- fail
Timeout Double - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover
Criterias List<String> - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- max
Fails Double - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read
Timeout Double - Default: 60000
- retries Double
- The number of retries to execute upon failure to proxy. Default: 5
- slots Double
- The number of slots in the load balancer algorithm. Default: 10000
- tokens
Count StringStrategy - Methodology to use for token usage calculation. possible known values include one of ["completion-tokens", "cost", "llm-accuracy", "prompt-tokens", "total-tokens"]; Default: "total-tokens"
- write
Timeout Double - Default: 60000
- connect
Timeout number - Default: 60000
- fail
Timeout number - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover
Criterias string[] - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- max
Fails number - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read
Timeout number - Default: 60000
- retries number
- The number of retries to execute upon failure to proxy. Default: 5
- slots number
- The number of slots in the load balancer algorithm. Default: 10000
- tokens
Count stringStrategy - Methodology to use for token usage calculation. possible known values include one of ["completion-tokens", "cost", "llm-accuracy", "prompt-tokens", "total-tokens"]; Default: "total-tokens"
- write
Timeout number - Default: 60000
- connect_
timeout float - Default: 60000
- fail_
timeout float - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover_
criterias Sequence[str] - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- max_
fails float - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read_
timeout float - Default: 60000
- retries float
- The number of retries to execute upon failure to proxy. Default: 5
- slots float
- The number of slots in the load balancer algorithm. Default: 10000
- tokens_
count_ strstrategy - Methodology to use for token usage calculation. possible known values include one of ["completion-tokens", "cost", "llm-accuracy", "prompt-tokens", "total-tokens"]; Default: "total-tokens"
- write_
timeout float - Default: 60000
- connect
Timeout Number - Default: 60000
- fail
Timeout Number - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover
Criterias List<String> - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- max
Fails Number - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read
Timeout Number - Default: 60000
- retries Number
- The number of retries to execute upon failure to proxy. Default: 5
- slots Number
- The number of slots in the load balancer algorithm. Default: 10000
- tokens
Count StringStrategy - Methodology to use for token usage calculation. possible known values include one of ["completion-tokens", "cost", "llm-accuracy", "prompt-tokens", "total-tokens"]; Default: "total-tokens"
- write
Timeout Number - Default: 60000
AiGatewayModelApiConfigBalancerPriority, AiGatewayModelApiConfigBalancerPriorityArgs
- Connect
Timeout double - Default: 60000
- Fail
Timeout double - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - Failover
Criterias List<string> - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- Max
Fails double - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - Read
Timeout double - Default: 60000
- Retries double
- The number of retries to execute upon failure to proxy. Default: 5
- Slots double
- The number of slots in the load balancer algorithm. Default: 10000
- Write
Timeout double - Default: 60000
- Connect
Timeout float64 - Default: 60000
- Fail
Timeout float64 - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - Failover
Criterias []string - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- Max
Fails float64 - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - Read
Timeout float64 - Default: 60000
- Retries float64
- The number of retries to execute upon failure to proxy. Default: 5
- Slots float64
- The number of slots in the load balancer algorithm. Default: 10000
- Write
Timeout float64 - Default: 60000
- connect_
timeout number - Default: 60000
- fail_
timeout number - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover_
criterias list(string) - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- max_
fails number - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read_
timeout number - Default: 60000
- retries number
- The number of retries to execute upon failure to proxy. Default: 5
- slots number
- The number of slots in the load balancer algorithm. Default: 10000
- write_
timeout number - Default: 60000
- connect
Timeout Double - Default: 60000
- fail
Timeout Double - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover
Criterias List<String> - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- max
Fails Double - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read
Timeout Double - Default: 60000
- retries Double
- The number of retries to execute upon failure to proxy. Default: 5
- slots Double
- The number of slots in the load balancer algorithm. Default: 10000
- write
Timeout Double - Default: 60000
- connect
Timeout number - Default: 60000
- fail
Timeout number - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover
Criterias string[] - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- max
Fails number - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read
Timeout number - Default: 60000
- retries number
- The number of retries to execute upon failure to proxy. Default: 5
- slots number
- The number of slots in the load balancer algorithm. Default: 10000
- write
Timeout number - Default: 60000
- connect_
timeout float - Default: 60000
- fail_
timeout float - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover_
criterias Sequence[str] - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- max_
fails float - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read_
timeout float - Default: 60000
- retries float
- The number of retries to execute upon failure to proxy. Default: 5
- slots float
- The number of slots in the load balancer algorithm. Default: 10000
- write_
timeout float - Default: 60000
- connect
Timeout Number - Default: 60000
- fail
Timeout Number - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover
Criterias List<String> - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- max
Fails Number - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read
Timeout Number - Default: 60000
- retries Number
- The number of retries to execute upon failure to proxy. Default: 5
- slots Number
- The number of slots in the load balancer algorithm. Default: 10000
- write
Timeout Number - Default: 60000
AiGatewayModelApiConfigBalancerRoundRobin, AiGatewayModelApiConfigBalancerRoundRobinArgs
- Connect
Timeout double - Default: 60000
- Fail
Timeout double - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - Failover
Criterias List<string> - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- Max
Fails double - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - Read
Timeout double - Default: 60000
- Retries double
- The number of retries to execute upon failure to proxy. Default: 5
- Slots double
- The number of slots in the load balancer algorithm. Default: 10000
- Write
Timeout double - Default: 60000
- Connect
Timeout float64 - Default: 60000
- Fail
Timeout float64 - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - Failover
Criterias []string - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- Max
Fails float64 - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - Read
Timeout float64 - Default: 60000
- Retries float64
- The number of retries to execute upon failure to proxy. Default: 5
- Slots float64
- The number of slots in the load balancer algorithm. Default: 10000
- Write
Timeout float64 - Default: 60000
- connect_
timeout number - Default: 60000
- fail_
timeout number - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover_
criterias list(string) - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- max_
fails number - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read_
timeout number - Default: 60000
- retries number
- The number of retries to execute upon failure to proxy. Default: 5
- slots number
- The number of slots in the load balancer algorithm. Default: 10000
- write_
timeout number - Default: 60000
- connect
Timeout Double - Default: 60000
- fail
Timeout Double - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover
Criterias List<String> - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- max
Fails Double - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read
Timeout Double - Default: 60000
- retries Double
- The number of retries to execute upon failure to proxy. Default: 5
- slots Double
- The number of slots in the load balancer algorithm. Default: 10000
- write
Timeout Double - Default: 60000
- connect
Timeout number - Default: 60000
- fail
Timeout number - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover
Criterias string[] - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- max
Fails number - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read
Timeout number - Default: 60000
- retries number
- The number of retries to execute upon failure to proxy. Default: 5
- slots number
- The number of slots in the load balancer algorithm. Default: 10000
- write
Timeout number - Default: 60000
- connect_
timeout float - Default: 60000
- fail_
timeout float - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover_
criterias Sequence[str] - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- max_
fails float - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read_
timeout float - Default: 60000
- retries float
- The number of retries to execute upon failure to proxy. Default: 5
- slots float
- The number of slots in the load balancer algorithm. Default: 10000
- write_
timeout float - Default: 60000
- connect
Timeout Number - Default: 60000
- fail
Timeout Number - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover
Criterias List<String> - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- max
Fails Number - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read
Timeout Number - Default: 60000
- retries Number
- The number of retries to execute upon failure to proxy. Default: 5
- slots Number
- The number of slots in the load balancer algorithm. Default: 10000
- write
Timeout Number - Default: 60000
AiGatewayModelApiConfigBalancerSemantic, AiGatewayModelApiConfigBalancerSemanticArgs
- Connect
Timeout double - Default: 60000
- Embeddings
Ai
Gateway Model Api Config Balancer Semantic Embeddings - Embeddings model configuration for this model. Not Null
- Fail
Timeout double - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - Failover
Criterias List<string> - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- Max
Fails double - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - Read
Timeout double - Default: 60000
- Retries double
- The number of retries to execute upon failure to proxy. Default: 5
- Slots double
- The number of slots in the load balancer algorithm. Default: 10000
- Vectordb
Ai
Gateway Model Api Config Balancer Semantic Vectordb - Configuration for the vector database used by the model. Not Null
- Write
Timeout double - Default: 60000
- Connect
Timeout float64 - Default: 60000
- Embeddings
Ai
Gateway Model Api Config Balancer Semantic Embeddings - Embeddings model configuration for this model. Not Null
- Fail
Timeout float64 - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - Failover
Criterias []string - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- Max
Fails float64 - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - Read
Timeout float64 - Default: 60000
- Retries float64
- The number of retries to execute upon failure to proxy. Default: 5
- Slots float64
- The number of slots in the load balancer algorithm. Default: 10000
- Vectordb
Ai
Gateway Model Api Config Balancer Semantic Vectordb - Configuration for the vector database used by the model. Not Null
- Write
Timeout float64 - Default: 60000
- connect_
timeout number - Default: 60000
- embeddings object
- Embeddings model configuration for this model. Not Null
- fail_
timeout number - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover_
criterias list(string) - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- max_
fails number - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read_
timeout number - Default: 60000
- retries number
- The number of retries to execute upon failure to proxy. Default: 5
- slots number
- The number of slots in the load balancer algorithm. Default: 10000
- vectordb object
- Configuration for the vector database used by the model. Not Null
- write_
timeout number - Default: 60000
- connect
Timeout Double - Default: 60000
- embeddings
Ai
Gateway Model Api Config Balancer Semantic Embeddings - Embeddings model configuration for this model. Not Null
- fail
Timeout Double - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover
Criterias List<String> - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- max
Fails Double - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read
Timeout Double - Default: 60000
- retries Double
- The number of retries to execute upon failure to proxy. Default: 5
- slots Double
- The number of slots in the load balancer algorithm. Default: 10000
- vectordb
Ai
Gateway Model Api Config Balancer Semantic Vectordb - Configuration for the vector database used by the model. Not Null
- write
Timeout Double - Default: 60000
- connect
Timeout number - Default: 60000
- embeddings
Ai
Gateway Model Api Config Balancer Semantic Embeddings - Embeddings model configuration for this model. Not Null
- fail
Timeout number - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover
Criterias string[] - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- max
Fails number - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read
Timeout number - Default: 60000
- retries number
- The number of retries to execute upon failure to proxy. Default: 5
- slots number
- The number of slots in the load balancer algorithm. Default: 10000
- vectordb
Ai
Gateway Model Api Config Balancer Semantic Vectordb - Configuration for the vector database used by the model. Not Null
- write
Timeout number - Default: 60000
- connect_
timeout float - Default: 60000
- embeddings
Ai
Gateway Model Api Config Balancer Semantic Embeddings - Embeddings model configuration for this model. Not Null
- fail_
timeout float - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover_
criterias Sequence[str] - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- max_
fails float - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read_
timeout float - Default: 60000
- retries float
- The number of retries to execute upon failure to proxy. Default: 5
- slots float
- The number of slots in the load balancer algorithm. Default: 10000
- vectordb
Ai
Gateway Model Api Config Balancer Semantic Vectordb - Configuration for the vector database used by the model. Not Null
- write_
timeout float - Default: 60000
- connect
Timeout Number - Default: 60000
- embeddings Property Map
- Embeddings model configuration for this model. Not Null
- fail
Timeout Number - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover
Criterias List<String> - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- max
Fails Number - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read
Timeout Number - Default: 60000
- retries Number
- The number of retries to execute upon failure to proxy. Default: 5
- slots Number
- The number of slots in the load balancer algorithm. Default: 10000
- vectordb Property Map
- Configuration for the vector database used by the model. Not Null
- write
Timeout Number - Default: 60000
AiGatewayModelApiConfigBalancerSemanticEmbeddings, AiGatewayModelApiConfigBalancerSemanticEmbeddingsArgs
- Allow
Auth boolOverride - When enabled, request-level auth parameters (such as API keys or bearer tokens) will override the static values defined for the provider. Default: false
- Config
Ai
Gateway Model Api Config Balancer Semantic Embeddings Config - Configuration for an embeddings model. Not Null
- Name string
- The name of the embeddings model. Not Null
- Provider string
- Reference to a model provider instance by name. Not Null
- Allow
Auth boolOverride - When enabled, request-level auth parameters (such as API keys or bearer tokens) will override the static values defined for the provider. Default: false
- Config
Ai
Gateway Model Api Config Balancer Semantic Embeddings Config - Configuration for an embeddings model. Not Null
- Name string
- The name of the embeddings model. Not Null
- Provider string
- Reference to a model provider instance by name. Not Null
- allow_
auth_ booloverride - When enabled, request-level auth parameters (such as API keys or bearer tokens) will override the static values defined for the provider. Default: false
- config object
- Configuration for an embeddings model. Not Null
- name string
- The name of the embeddings model. Not Null
- provider string
- Reference to a model provider instance by name. Not Null
- allow
Auth BooleanOverride - When enabled, request-level auth parameters (such as API keys or bearer tokens) will override the static values defined for the provider. Default: false
- config
Ai
Gateway Model Api Config Balancer Semantic Embeddings Config - Configuration for an embeddings model. Not Null
- name String
- The name of the embeddings model. Not Null
- provider String
- Reference to a model provider instance by name. Not Null
- allow
Auth booleanOverride - When enabled, request-level auth parameters (such as API keys or bearer tokens) will override the static values defined for the provider. Default: false
- config
Ai
Gateway Model Api Config Balancer Semantic Embeddings Config - Configuration for an embeddings model. Not Null
- name string
- The name of the embeddings model. Not Null
- provider string
- Reference to a model provider instance by name. Not Null
- allow_
auth_ booloverride - When enabled, request-level auth parameters (such as API keys or bearer tokens) will override the static values defined for the provider. Default: false
- config
Ai
Gateway Model Api Config Balancer Semantic Embeddings Config - Configuration for an embeddings model. Not Null
- name str
- The name of the embeddings model. Not Null
- provider str
- Reference to a model provider instance by name. Not Null
- allow
Auth BooleanOverride - When enabled, request-level auth parameters (such as API keys or bearer tokens) will override the static values defined for the provider. Default: false
- config Property Map
- Configuration for an embeddings model. Not Null
- name String
- The name of the embeddings model. Not Null
- provider String
- Reference to a model provider instance by name. Not Null
AiGatewayModelApiConfigBalancerSemanticEmbeddingsConfig, AiGatewayModelApiConfigBalancerSemanticEmbeddingsConfigArgs
- Azure
Ai
Gateway Model Api Config Balancer Semantic Embeddings Config Azure - Azure OpenAI-specific configuration for an embeddings model. Azure AI Foundry embeddings are not supported.
- Bedrock
Ai
Gateway Model Api Config Balancer Semantic Embeddings Config Bedrock - AWS Bedrock-specific configuration for a model.
- Gemini
Ai
Gateway Model Api Config Balancer Semantic Embeddings Config Gemini - Google Gemini-specific configuration for a model.
- Huggingface
Ai
Gateway Model Api Config Balancer Semantic Embeddings Config Huggingface - Hugging Face-specific configuration for a model.
- Mistral
Ai
Gateway Model Api Config Balancer Semantic Embeddings Config Mistral - Ollama
Ai
Gateway Model Api Config Balancer Semantic Embeddings Config Ollama - Ollama-specific configuration for a model.
- Openai
Ai
Gateway Model Api Config Balancer Semantic Embeddings Config Openai - Openai-specific configuration for a model.
- Azure
Ai
Gateway Model Api Config Balancer Semantic Embeddings Config Azure - Azure OpenAI-specific configuration for an embeddings model. Azure AI Foundry embeddings are not supported.
- Bedrock
Ai
Gateway Model Api Config Balancer Semantic Embeddings Config Bedrock - AWS Bedrock-specific configuration for a model.
- Gemini
Ai
Gateway Model Api Config Balancer Semantic Embeddings Config Gemini - Google Gemini-specific configuration for a model.
- Huggingface
Ai
Gateway Model Api Config Balancer Semantic Embeddings Config Huggingface - Hugging Face-specific configuration for a model.
- Mistral
Ai
Gateway Model Api Config Balancer Semantic Embeddings Config Mistral - Ollama
Ai
Gateway Model Api Config Balancer Semantic Embeddings Config Ollama - Ollama-specific configuration for a model.
- Openai
Ai
Gateway Model Api Config Balancer Semantic Embeddings Config Openai - Openai-specific configuration for a model.
- azure object
- Azure OpenAI-specific configuration for an embeddings model. Azure AI Foundry embeddings are not supported.
- bedrock object
- AWS Bedrock-specific configuration for a model.
- gemini object
- Google Gemini-specific configuration for a model.
- huggingface object
- Hugging Face-specific configuration for a model.
- mistral object
- ollama object
- Ollama-specific configuration for a model.
- openai object
- Openai-specific configuration for a model.
- azure
Ai
Gateway Model Api Config Balancer Semantic Embeddings Config Azure - Azure OpenAI-specific configuration for an embeddings model. Azure AI Foundry embeddings are not supported.
- bedrock
Ai
Gateway Model Api Config Balancer Semantic Embeddings Config Bedrock - AWS Bedrock-specific configuration for a model.
- gemini
Ai
Gateway Model Api Config Balancer Semantic Embeddings Config Gemini - Google Gemini-specific configuration for a model.
- huggingface
Ai
Gateway Model Api Config Balancer Semantic Embeddings Config Huggingface - Hugging Face-specific configuration for a model.
- mistral
Ai
Gateway Model Api Config Balancer Semantic Embeddings Config Mistral - ollama
Ai
Gateway Model Api Config Balancer Semantic Embeddings Config Ollama - Ollama-specific configuration for a model.
- openai
Ai
Gateway Model Api Config Balancer Semantic Embeddings Config Openai - Openai-specific configuration for a model.
- azure
Ai
Gateway Model Api Config Balancer Semantic Embeddings Config Azure - Azure OpenAI-specific configuration for an embeddings model. Azure AI Foundry embeddings are not supported.
- bedrock
Ai
Gateway Model Api Config Balancer Semantic Embeddings Config Bedrock - AWS Bedrock-specific configuration for a model.
- gemini
Ai
Gateway Model Api Config Balancer Semantic Embeddings Config Gemini - Google Gemini-specific configuration for a model.
- huggingface
Ai
Gateway Model Api Config Balancer Semantic Embeddings Config Huggingface - Hugging Face-specific configuration for a model.
- mistral
Ai
Gateway Model Api Config Balancer Semantic Embeddings Config Mistral - ollama
Ai
Gateway Model Api Config Balancer Semantic Embeddings Config Ollama - Ollama-specific configuration for a model.
- openai
Ai
Gateway Model Api Config Balancer Semantic Embeddings Config Openai - Openai-specific configuration for a model.
- azure
Ai
Gateway Model Api Config Balancer Semantic Embeddings Config Azure - Azure OpenAI-specific configuration for an embeddings model. Azure AI Foundry embeddings are not supported.
- bedrock
Ai
Gateway Model Api Config Balancer Semantic Embeddings Config Bedrock - AWS Bedrock-specific configuration for a model.
- gemini
Ai
Gateway Model Api Config Balancer Semantic Embeddings Config Gemini - Google Gemini-specific configuration for a model.
- huggingface
Ai
Gateway Model Api Config Balancer Semantic Embeddings Config Huggingface - Hugging Face-specific configuration for a model.
- mistral
Ai
Gateway Model Api Config Balancer Semantic Embeddings Config Mistral - ollama
Ai
Gateway Model Api Config Balancer Semantic Embeddings Config Ollama - Ollama-specific configuration for a model.
- openai
Ai
Gateway Model Api Config Balancer Semantic Embeddings Config Openai - Openai-specific configuration for a model.
- azure Property Map
- Azure OpenAI-specific configuration for an embeddings model. Azure AI Foundry embeddings are not supported.
- bedrock Property Map
- AWS Bedrock-specific configuration for a model.
- gemini Property Map
- Google Gemini-specific configuration for a model.
- huggingface Property Map
- Hugging Face-specific configuration for a model.
- mistral Property Map
- ollama Property Map
- Ollama-specific configuration for a model.
- openai Property Map
- Openai-specific configuration for a model.
AiGatewayModelApiConfigBalancerSemanticEmbeddingsConfigAzure, AiGatewayModelApiConfigBalancerSemanticEmbeddingsConfigAzureArgs
- Api
Version string - The Azure OpenAI API version to use. Default: "2023-05-15"
- Deployment
Id string - The Azure OpenAI deployment ID for the embeddings model. Not Null
- Type string
- Not Null; must be "azure"
- Upstream
Url string - The URL of the embeddings model.
- Api
Version string - The Azure OpenAI API version to use. Default: "2023-05-15"
- Deployment
Id string - The Azure OpenAI deployment ID for the embeddings model. Not Null
- Type string
- Not Null; must be "azure"
- Upstream
Url string - The URL of the embeddings model.
- api_
version string - The Azure OpenAI API version to use. Default: "2023-05-15"
- deployment_
id string - The Azure OpenAI deployment ID for the embeddings model. Not Null
- type string
- Not Null; must be "azure"
- upstream_
url string - The URL of the embeddings model.
- api
Version String - The Azure OpenAI API version to use. Default: "2023-05-15"
- deployment
Id String - The Azure OpenAI deployment ID for the embeddings model. Not Null
- type String
- Not Null; must be "azure"
- upstream
Url String - The URL of the embeddings model.
- api
Version string - The Azure OpenAI API version to use. Default: "2023-05-15"
- deployment
Id string - The Azure OpenAI deployment ID for the embeddings model. Not Null
- type string
- Not Null; must be "azure"
- upstream
Url string - The URL of the embeddings model.
- api_
version str - The Azure OpenAI API version to use. Default: "2023-05-15"
- deployment_
id str - The Azure OpenAI deployment ID for the embeddings model. Not Null
- type str
- Not Null; must be "azure"
- upstream_
url str - The URL of the embeddings model.
- api
Version String - The Azure OpenAI API version to use. Default: "2023-05-15"
- deployment
Id String - The Azure OpenAI deployment ID for the embeddings model. Not Null
- type String
- Not Null; must be "azure"
- upstream
Url String - The URL of the embeddings model.
AiGatewayModelApiConfigBalancerSemanticEmbeddingsConfigBedrock, AiGatewayModelApiConfigBalancerSemanticEmbeddingsConfigBedrockArgs
- Batch
Bucket stringPrefix - S3 bucket prefix for batch inference jobs.
- Embeddings
Normalize bool - Whether to normalize embedding vectors in the response. Default: false
- Performance
Config stringLatency - Latency performance configuration for the model invocation.
- Region string
- The AWS region for the model. Setting this option overrides the AWS_REGION environment variable.
- Upstream
Url string - The URL of the embeddings model.
- Video
Output stringS3Uri - S3 URI for storing video generation outputs.
- Batch
Bucket stringPrefix - S3 bucket prefix for batch inference jobs.
- Embeddings
Normalize bool - Whether to normalize embedding vectors in the response. Default: false
- Performance
Config stringLatency - Latency performance configuration for the model invocation.
- Region string
- The AWS region for the model. Setting this option overrides the AWS_REGION environment variable.
- Upstream
Url string - The URL of the embeddings model.
- Video
Output stringS3Uri - S3 URI for storing video generation outputs.
- batch_
bucket_ stringprefix - S3 bucket prefix for batch inference jobs.
- embeddings_
normalize bool - Whether to normalize embedding vectors in the response. Default: false
- performance_
config_ stringlatency - Latency performance configuration for the model invocation.
- region string
- The AWS region for the model. Setting this option overrides the AWS_REGION environment variable.
- upstream_
url string - The URL of the embeddings model.
- video_
output_ strings3_ uri - S3 URI for storing video generation outputs.
- batch
Bucket StringPrefix - S3 bucket prefix for batch inference jobs.
- embeddings
Normalize Boolean - Whether to normalize embedding vectors in the response. Default: false
- performance
Config StringLatency - Latency performance configuration for the model invocation.
- region String
- The AWS region for the model. Setting this option overrides the AWS_REGION environment variable.
- upstream
Url String - The URL of the embeddings model.
- video
Output StringS3Uri - S3 URI for storing video generation outputs.
- batch
Bucket stringPrefix - S3 bucket prefix for batch inference jobs.
- embeddings
Normalize boolean - Whether to normalize embedding vectors in the response. Default: false
- performance
Config stringLatency - Latency performance configuration for the model invocation.
- region string
- The AWS region for the model. Setting this option overrides the AWS_REGION environment variable.
- upstream
Url string - The URL of the embeddings model.
- video
Output stringS3Uri - S3 URI for storing video generation outputs.
- batch_
bucket_ strprefix - S3 bucket prefix for batch inference jobs.
- embeddings_
normalize bool - Whether to normalize embedding vectors in the response. Default: false
- performance_
config_ strlatency - Latency performance configuration for the model invocation.
- region str
- The AWS region for the model. Setting this option overrides the AWS_REGION environment variable.
- upstream_
url str - The URL of the embeddings model.
- video_
output_ strs3_ uri - S3 URI for storing video generation outputs.
- batch
Bucket StringPrefix - S3 bucket prefix for batch inference jobs.
- embeddings
Normalize Boolean - Whether to normalize embedding vectors in the response. Default: false
- performance
Config StringLatency - Latency performance configuration for the model invocation.
- region String
- The AWS region for the model. Setting this option overrides the AWS_REGION environment variable.
- upstream
Url String - The URL of the embeddings model.
- video
Output StringS3Uri - S3 URI for storing video generation outputs.
AiGatewayModelApiConfigBalancerSemanticEmbeddingsConfigGemini, AiGatewayModelApiConfigBalancerSemanticEmbeddingsConfigGeminiArgs
- Gcp
Environment AiGateway Model Api Config Balancer Semantic Embeddings Config Gemini Gcp Environment - Configuration for a model hosted on Google Cloud Project.
- Upstream
Url string - The URL of the embeddings model.
- Gcp
Environment AiGateway Model Api Config Balancer Semantic Embeddings Config Gemini Gcp Environment - Configuration for a model hosted on Google Cloud Project.
- Upstream
Url string - The URL of the embeddings model.
- gcp_
environment object - Configuration for a model hosted on Google Cloud Project.
- upstream_
url string - The URL of the embeddings model.
- gcp
Environment AiGateway Model Api Config Balancer Semantic Embeddings Config Gemini Gcp Environment - Configuration for a model hosted on Google Cloud Project.
- upstream
Url String - The URL of the embeddings model.
- gcp
Environment AiGateway Model Api Config Balancer Semantic Embeddings Config Gemini Gcp Environment - Configuration for a model hosted on Google Cloud Project.
- upstream
Url string - The URL of the embeddings model.
- gcp_
environment AiGateway Model Api Config Balancer Semantic Embeddings Config Gemini Gcp Environment - Configuration for a model hosted on Google Cloud Project.
- upstream_
url str - The URL of the embeddings model.
- gcp
Environment Property Map - Configuration for a model hosted on Google Cloud Project.
- upstream
Url String - The URL of the embeddings model.
AiGatewayModelApiConfigBalancerSemanticEmbeddingsConfigGeminiGcpEnvironment, AiGatewayModelApiConfigBalancerSemanticEmbeddingsConfigGeminiGcpEnvironmentArgs
- Api
Endpoint string - The custom API endpoint for the Gemini model. Not Null
- Location
Id string - The Google Cloud location ID for the model endpoint. Not Null
- Project
Id string - The Google Cloud project ID for the model endpoint. Not Null
- Api
Endpoint string - The custom API endpoint for the Gemini model. Not Null
- Location
Id string - The Google Cloud location ID for the model endpoint. Not Null
- Project
Id string - The Google Cloud project ID for the model endpoint. Not Null
- api_
endpoint string - The custom API endpoint for the Gemini model. Not Null
- location_
id string - The Google Cloud location ID for the model endpoint. Not Null
- project_
id string - The Google Cloud project ID for the model endpoint. Not Null
- api
Endpoint String - The custom API endpoint for the Gemini model. Not Null
- location
Id String - The Google Cloud location ID for the model endpoint. Not Null
- project
Id String - The Google Cloud project ID for the model endpoint. Not Null
- api
Endpoint string - The custom API endpoint for the Gemini model. Not Null
- location
Id string - The Google Cloud location ID for the model endpoint. Not Null
- project
Id string - The Google Cloud project ID for the model endpoint. Not Null
- api_
endpoint str - The custom API endpoint for the Gemini model. Not Null
- location_
id str - The Google Cloud location ID for the model endpoint. Not Null
- project_
id str - The Google Cloud project ID for the model endpoint. Not Null
- api
Endpoint String - The custom API endpoint for the Gemini model. Not Null
- location
Id String - The Google Cloud location ID for the model endpoint. Not Null
- project
Id String - The Google Cloud project ID for the model endpoint. Not Null
AiGatewayModelApiConfigBalancerSemanticEmbeddingsConfigHuggingface, AiGatewayModelApiConfigBalancerSemanticEmbeddingsConfigHuggingfaceArgs
- Upstream
Url string - The URL of the embeddings model.
- Use
Cache bool - Whether to use the Hugging Face inference cache. Default: false
- Wait
For boolModel - Whether to wait for the model to load if it is not ready. Default: false
- Upstream
Url string - The URL of the embeddings model.
- Use
Cache bool - Whether to use the Hugging Face inference cache. Default: false
- Wait
For boolModel - Whether to wait for the model to load if it is not ready. Default: false
- upstream_
url string - The URL of the embeddings model.
- use_
cache bool - Whether to use the Hugging Face inference cache. Default: false
- wait_
for_ boolmodel - Whether to wait for the model to load if it is not ready. Default: false
- upstream
Url String - The URL of the embeddings model.
- use
Cache Boolean - Whether to use the Hugging Face inference cache. Default: false
- wait
For BooleanModel - Whether to wait for the model to load if it is not ready. Default: false
- upstream
Url string - The URL of the embeddings model.
- use
Cache boolean - Whether to use the Hugging Face inference cache. Default: false
- wait
For booleanModel - Whether to wait for the model to load if it is not ready. Default: false
- upstream_
url str - The URL of the embeddings model.
- use_
cache bool - Whether to use the Hugging Face inference cache. Default: false
- wait_
for_ boolmodel - Whether to wait for the model to load if it is not ready. Default: false
- upstream
Url String - The URL of the embeddings model.
- use
Cache Boolean - Whether to use the Hugging Face inference cache. Default: false
- wait
For BooleanModel - Whether to wait for the model to load if it is not ready. Default: false
AiGatewayModelApiConfigBalancerSemanticEmbeddingsConfigMistral, AiGatewayModelApiConfigBalancerSemanticEmbeddingsConfigMistralArgs
- Type string
- Not Null; must be "mistral"
- Upstream
Url string - The URL of the embeddings model.
- Type string
- Not Null; must be "mistral"
- Upstream
Url string - The URL of the embeddings model.
- type string
- Not Null; must be "mistral"
- upstream_
url string - The URL of the embeddings model.
- type String
- Not Null; must be "mistral"
- upstream
Url String - The URL of the embeddings model.
- type string
- Not Null; must be "mistral"
- upstream
Url string - The URL of the embeddings model.
- type str
- Not Null; must be "mistral"
- upstream_
url str - The URL of the embeddings model.
- type String
- Not Null; must be "mistral"
- upstream
Url String - The URL of the embeddings model.
AiGatewayModelApiConfigBalancerSemanticEmbeddingsConfigOllama, AiGatewayModelApiConfigBalancerSemanticEmbeddingsConfigOllamaArgs
- Upstream
Url string - The URL of the embeddings model.
- Upstream
Url string - The URL of the embeddings model.
- upstream_
url string - The URL of the embeddings model.
- upstream
Url String - The URL of the embeddings model.
- upstream
Url string - The URL of the embeddings model.
- upstream_
url str - The URL of the embeddings model.
- upstream
Url String - The URL of the embeddings model.
AiGatewayModelApiConfigBalancerSemanticEmbeddingsConfigOpenai, AiGatewayModelApiConfigBalancerSemanticEmbeddingsConfigOpenaiArgs
- Upstream
Url string - The URL of the embeddings model.
- Upstream
Url string - The URL of the embeddings model.
- upstream_
url string - The URL of the embeddings model.
- upstream
Url String - The URL of the embeddings model.
- upstream
Url string - The URL of the embeddings model.
- upstream_
url str - The URL of the embeddings model.
- upstream
Url String - The URL of the embeddings model.
AiGatewayModelApiConfigBalancerSemanticVectordb, AiGatewayModelApiConfigBalancerSemanticVectordbArgs
- Pgvector
Ai
Gateway Model Api Config Balancer Semantic Vectordb Pgvector - Redis
Ai
Gateway Model Api Config Balancer Semantic Vectordb Redis - Config for connecting to a Cloud Provider's Redis instance.
- Pgvector
Ai
Gateway Model Api Config Balancer Semantic Vectordb Pgvector - Redis
Ai
Gateway Model Api Config Balancer Semantic Vectordb Redis - Config for connecting to a Cloud Provider's Redis instance.
- pgvector
Ai
Gateway Model Api Config Balancer Semantic Vectordb Pgvector - redis
Ai
Gateway Model Api Config Balancer Semantic Vectordb Redis - Config for connecting to a Cloud Provider's Redis instance.
- pgvector
Ai
Gateway Model Api Config Balancer Semantic Vectordb Pgvector - redis
Ai
Gateway Model Api Config Balancer Semantic Vectordb Redis - Config for connecting to a Cloud Provider's Redis instance.
- pgvector
Ai
Gateway Model Api Config Balancer Semantic Vectordb Pgvector - redis
Ai
Gateway Model Api Config Balancer Semantic Vectordb Redis - Config for connecting to a Cloud Provider's Redis instance.
- pgvector Property Map
- redis Property Map
- Config for connecting to a Cloud Provider's Redis instance.
AiGatewayModelApiConfigBalancerSemanticVectordbPgvector, AiGatewayModelApiConfigBalancerSemanticVectordbPgvectorArgs
- Database string
- the database of the pgvector database. Default: "kong-pgvector"
- Dimensions double
- the desired dimensionality for the vectors. Not Null
- Distance
Metric string - the distance metric to use for vector searches. possible known values include one of ["cosine", "euclidean"]; Not Null
- Host string
- the host of the pgvector database. Default: "127.0.0.1"
- Password string
- the password of the pgvector database This field is referenceable.
- Port double
- the port of the pgvector database. Default: 5432
- Ssl
Ai
Gateway Model Api Config Balancer Semantic Vectordb Pgvector Ssl - Threshold double
- the default similarity threshold for accepting semantic search results (float). Higher threshold means more results are considered similar.
- Timeout double
- the timeout of the pgvector database. Default: 5000
- User string
- the user of the pgvector database This field is referenceable. Default: "postgres"
- Database string
- the database of the pgvector database. Default: "kong-pgvector"
- Dimensions float64
- the desired dimensionality for the vectors. Not Null
- Distance
Metric string - the distance metric to use for vector searches. possible known values include one of ["cosine", "euclidean"]; Not Null
- Host string
- the host of the pgvector database. Default: "127.0.0.1"
- Password string
- the password of the pgvector database This field is referenceable.
- Port float64
- the port of the pgvector database. Default: 5432
- Ssl
Ai
Gateway Model Api Config Balancer Semantic Vectordb Pgvector Ssl - Threshold float64
- the default similarity threshold for accepting semantic search results (float). Higher threshold means more results are considered similar.
- Timeout float64
- the timeout of the pgvector database. Default: 5000
- User string
- the user of the pgvector database This field is referenceable. Default: "postgres"
- database string
- the database of the pgvector database. Default: "kong-pgvector"
- dimensions number
- the desired dimensionality for the vectors. Not Null
- distance_
metric string - the distance metric to use for vector searches. possible known values include one of ["cosine", "euclidean"]; Not Null
- host string
- the host of the pgvector database. Default: "127.0.0.1"
- password string
- the password of the pgvector database This field is referenceable.
- port number
- the port of the pgvector database. Default: 5432
- ssl object
- threshold number
- the default similarity threshold for accepting semantic search results (float). Higher threshold means more results are considered similar.
- timeout number
- the timeout of the pgvector database. Default: 5000
- user string
- the user of the pgvector database This field is referenceable. Default: "postgres"
- database String
- the database of the pgvector database. Default: "kong-pgvector"
- dimensions Double
- the desired dimensionality for the vectors. Not Null
- distance
Metric String - the distance metric to use for vector searches. possible known values include one of ["cosine", "euclidean"]; Not Null
- host String
- the host of the pgvector database. Default: "127.0.0.1"
- password String
- the password of the pgvector database This field is referenceable.
- port Double
- the port of the pgvector database. Default: 5432
- ssl
Ai
Gateway Model Api Config Balancer Semantic Vectordb Pgvector Ssl - threshold Double
- the default similarity threshold for accepting semantic search results (float). Higher threshold means more results are considered similar.
- timeout Double
- the timeout of the pgvector database. Default: 5000
- user String
- the user of the pgvector database This field is referenceable. Default: "postgres"
- database string
- the database of the pgvector database. Default: "kong-pgvector"
- dimensions number
- the desired dimensionality for the vectors. Not Null
- distance
Metric string - the distance metric to use for vector searches. possible known values include one of ["cosine", "euclidean"]; Not Null
- host string
- the host of the pgvector database. Default: "127.0.0.1"
- password string
- the password of the pgvector database This field is referenceable.
- port number
- the port of the pgvector database. Default: 5432
- ssl
Ai
Gateway Model Api Config Balancer Semantic Vectordb Pgvector Ssl - threshold number
- the default similarity threshold for accepting semantic search results (float). Higher threshold means more results are considered similar.
- timeout number
- the timeout of the pgvector database. Default: 5000
- user string
- the user of the pgvector database This field is referenceable. Default: "postgres"
- database str
- the database of the pgvector database. Default: "kong-pgvector"
- dimensions float
- the desired dimensionality for the vectors. Not Null
- distance_
metric str - the distance metric to use for vector searches. possible known values include one of ["cosine", "euclidean"]; Not Null
- host str
- the host of the pgvector database. Default: "127.0.0.1"
- password str
- the password of the pgvector database This field is referenceable.
- port float
- the port of the pgvector database. Default: 5432
- ssl
Ai
Gateway Model Api Config Balancer Semantic Vectordb Pgvector Ssl - threshold float
- the default similarity threshold for accepting semantic search results (float). Higher threshold means more results are considered similar.
- timeout float
- the timeout of the pgvector database. Default: 5000
- user str
- the user of the pgvector database This field is referenceable. Default: "postgres"
- database String
- the database of the pgvector database. Default: "kong-pgvector"
- dimensions Number
- the desired dimensionality for the vectors. Not Null
- distance
Metric String - the distance metric to use for vector searches. possible known values include one of ["cosine", "euclidean"]; Not Null
- host String
- the host of the pgvector database. Default: "127.0.0.1"
- password String
- the password of the pgvector database This field is referenceable.
- port Number
- the port of the pgvector database. Default: 5432
- ssl Property Map
- threshold Number
- the default similarity threshold for accepting semantic search results (float). Higher threshold means more results are considered similar.
- timeout Number
- the timeout of the pgvector database. Default: 5000
- user String
- the user of the pgvector database This field is referenceable. Default: "postgres"
AiGatewayModelApiConfigBalancerSemanticVectordbPgvectorSsl, AiGatewayModelApiConfigBalancerSemanticVectordbPgvectorSslArgs
- Cert string
- the path of ssl cert to use for the pgvector database
- Cert
Key string - the path of ssl cert key to use for the pgvector database
- Enabled bool
- whether to use ssl for the pgvector database. Default: true
- Required bool
- whether ssl is required for the pgvector database. Default: true
- Verify bool
- whether to verify ssl for the pgvector database. Default: true
- Version string
- the ssl version to use for the pgvector database. possible known values include one of ["any", "tlsv12", "tlsv13"]; Default: "tlsv1_2"
- Cert string
- the path of ssl cert to use for the pgvector database
- Cert
Key string - the path of ssl cert key to use for the pgvector database
- Enabled bool
- whether to use ssl for the pgvector database. Default: true
- Required bool
- whether ssl is required for the pgvector database. Default: true
- Verify bool
- whether to verify ssl for the pgvector database. Default: true
- Version string
- the ssl version to use for the pgvector database. possible known values include one of ["any", "tlsv12", "tlsv13"]; Default: "tlsv1_2"
- cert string
- the path of ssl cert to use for the pgvector database
- cert_
key string - the path of ssl cert key to use for the pgvector database
- enabled bool
- whether to use ssl for the pgvector database. Default: true
- required bool
- whether ssl is required for the pgvector database. Default: true
- verify bool
- whether to verify ssl for the pgvector database. Default: true
- version string
- the ssl version to use for the pgvector database. possible known values include one of ["any", "tlsv12", "tlsv13"]; Default: "tlsv1_2"
- cert String
- the path of ssl cert to use for the pgvector database
- cert
Key String - the path of ssl cert key to use for the pgvector database
- enabled Boolean
- whether to use ssl for the pgvector database. Default: true
- required Boolean
- whether ssl is required for the pgvector database. Default: true
- verify Boolean
- whether to verify ssl for the pgvector database. Default: true
- version String
- the ssl version to use for the pgvector database. possible known values include one of ["any", "tlsv12", "tlsv13"]; Default: "tlsv1_2"
- cert string
- the path of ssl cert to use for the pgvector database
- cert
Key string - the path of ssl cert key to use for the pgvector database
- enabled boolean
- whether to use ssl for the pgvector database. Default: true
- required boolean
- whether ssl is required for the pgvector database. Default: true
- verify boolean
- whether to verify ssl for the pgvector database. Default: true
- version string
- the ssl version to use for the pgvector database. possible known values include one of ["any", "tlsv12", "tlsv13"]; Default: "tlsv1_2"
- cert str
- the path of ssl cert to use for the pgvector database
- cert_
key str - the path of ssl cert key to use for the pgvector database
- enabled bool
- whether to use ssl for the pgvector database. Default: true
- required bool
- whether ssl is required for the pgvector database. Default: true
- verify bool
- whether to verify ssl for the pgvector database. Default: true
- version str
- the ssl version to use for the pgvector database. possible known values include one of ["any", "tlsv12", "tlsv13"]; Default: "tlsv1_2"
- cert String
- the path of ssl cert to use for the pgvector database
- cert
Key String - the path of ssl cert key to use for the pgvector database
- enabled Boolean
- whether to use ssl for the pgvector database. Default: true
- required Boolean
- whether ssl is required for the pgvector database. Default: true
- verify Boolean
- whether to verify ssl for the pgvector database. Default: true
- version String
- the ssl version to use for the pgvector database. possible known values include one of ["any", "tlsv12", "tlsv13"]; Default: "tlsv1_2"
AiGatewayModelApiConfigBalancerSemanticVectordbRedis, AiGatewayModelApiConfigBalancerSemanticVectordbRedisArgs
- Cloud
Authentication AiGateway Model Api Config Balancer Semantic Vectordb Redis Cloud Authentication - Auth related config for connecting to a Cloud Provider's Redis instance.
- Cluster
Ai
Gateway Model Api Config Balancer Semantic Vectordb Redis Cluster - Cluster configuration for the Redis connection.
- Connect
Timeout double - An integer representing a timeout in milliseconds. Must be between 0 and 2^31-2. Default: 2000
- Connection
Is boolProxied - If the connection to Redis is proxied (e.g. Envoy), set it
true. Set thehostandportto point to the proxy address. Default: false - Database double
- Database to use for the Redis connection when using the
redisstrategy. Default: 0 - Dimensions double
- the desired dimensionality for the vectors. Not Null
- Distance
Metric string - the distance metric to use for vector searches. possible known values include one of ["cosine", "euclidean"]; Not Null
- Host string
- A string representing a host name, such as example.com. This field is referenceable. Default: "127.0.0.1"
- Keepalive
Ai
Gateway Model Api Config Balancer Semantic Vectordb Redis Keepalive - Keepalive configuration for the Redis connection.
- Password string
- Password to use for Redis connections. If undefined, no AUTH commands are sent to Redis. This field is referenceable.
- Port string
- An integer representing a port number between 0 and 65535, inclusive. This field is referenceable. Default: "6379"
- Read
Timeout double - An integer representing a timeout in milliseconds. Must be between 0 and 2^31-2. Default: 2000
- Send
Timeout double - An integer representing a timeout in milliseconds. Must be between 0 and 2^31-2. Default: 2000
- Sentinel
Ai
Gateway Model Api Config Balancer Semantic Vectordb Redis Sentinel - Configuration for Redis Sentinel.
- Server
Name string - A string representing an SNI (server name indication) value for TLS. This field is referenceable.
- Ssl bool
- If set to true, uses SSL to connect to Redis. Default: true
- Ssl
Verify bool - If set to true, verifies the validity of the server SSL certificate. If setting this parameter, also configure
lua_ssl_trusted_certificateinkong.confto specify the CA (or server) certificate used by your Redis server. You may also need to configurelua_ssl_verify_depthaccordingly. Default: true - Threshold double
- the default similarity threshold for accepting semantic search results (float). Higher threshold means more results are considered similar.
- Username string
- Username to use for Redis connections. If undefined, ACL authentication won't be performed. This requires Redis v6.0.0+. To be compatible with Redis v5.x.y, you can set it to
default. This field is referenceable.
- Cloud
Authentication AiGateway Model Api Config Balancer Semantic Vectordb Redis Cloud Authentication - Auth related config for connecting to a Cloud Provider's Redis instance.
- Cluster
Ai
Gateway Model Api Config Balancer Semantic Vectordb Redis Cluster - Cluster configuration for the Redis connection.
- Connect
Timeout float64 - An integer representing a timeout in milliseconds. Must be between 0 and 2^31-2. Default: 2000
- Connection
Is boolProxied - If the connection to Redis is proxied (e.g. Envoy), set it
true. Set thehostandportto point to the proxy address. Default: false - Database float64
- Database to use for the Redis connection when using the
redisstrategy. Default: 0 - Dimensions float64
- the desired dimensionality for the vectors. Not Null
- Distance
Metric string - the distance metric to use for vector searches. possible known values include one of ["cosine", "euclidean"]; Not Null
- Host string
- A string representing a host name, such as example.com. This field is referenceable. Default: "127.0.0.1"
- Keepalive
Ai
Gateway Model Api Config Balancer Semantic Vectordb Redis Keepalive - Keepalive configuration for the Redis connection.
- Password string
- Password to use for Redis connections. If undefined, no AUTH commands are sent to Redis. This field is referenceable.
- Port string
- An integer representing a port number between 0 and 65535, inclusive. This field is referenceable. Default: "6379"
- Read
Timeout float64 - An integer representing a timeout in milliseconds. Must be between 0 and 2^31-2. Default: 2000
- Send
Timeout float64 - An integer representing a timeout in milliseconds. Must be between 0 and 2^31-2. Default: 2000
- Sentinel
Ai
Gateway Model Api Config Balancer Semantic Vectordb Redis Sentinel - Configuration for Redis Sentinel.
- Server
Name string - A string representing an SNI (server name indication) value for TLS. This field is referenceable.
- Ssl bool
- If set to true, uses SSL to connect to Redis. Default: true
- Ssl
Verify bool - If set to true, verifies the validity of the server SSL certificate. If setting this parameter, also configure
lua_ssl_trusted_certificateinkong.confto specify the CA (or server) certificate used by your Redis server. You may also need to configurelua_ssl_verify_depthaccordingly. Default: true - Threshold float64
- the default similarity threshold for accepting semantic search results (float). Higher threshold means more results are considered similar.
- Username string
- Username to use for Redis connections. If undefined, ACL authentication won't be performed. This requires Redis v6.0.0+. To be compatible with Redis v5.x.y, you can set it to
default. This field is referenceable.
- cloud_
authentication object - Auth related config for connecting to a Cloud Provider's Redis instance.
- cluster object
- Cluster configuration for the Redis connection.
- connect_
timeout number - An integer representing a timeout in milliseconds. Must be between 0 and 2^31-2. Default: 2000
- connection_
is_ boolproxied - If the connection to Redis is proxied (e.g. Envoy), set it
true. Set thehostandportto point to the proxy address. Default: false - database number
- Database to use for the Redis connection when using the
redisstrategy. Default: 0 - dimensions number
- the desired dimensionality for the vectors. Not Null
- distance_
metric string - the distance metric to use for vector searches. possible known values include one of ["cosine", "euclidean"]; Not Null
- host string
- A string representing a host name, such as example.com. This field is referenceable. Default: "127.0.0.1"
- keepalive object
- Keepalive configuration for the Redis connection.
- password string
- Password to use for Redis connections. If undefined, no AUTH commands are sent to Redis. This field is referenceable.
- port string
- An integer representing a port number between 0 and 65535, inclusive. This field is referenceable. Default: "6379"
- read_
timeout number - An integer representing a timeout in milliseconds. Must be between 0 and 2^31-2. Default: 2000
- send_
timeout number - An integer representing a timeout in milliseconds. Must be between 0 and 2^31-2. Default: 2000
- sentinel object
- Configuration for Redis Sentinel.
- server_
name string - A string representing an SNI (server name indication) value for TLS. This field is referenceable.
- ssl bool
- If set to true, uses SSL to connect to Redis. Default: true
- ssl_
verify bool - If set to true, verifies the validity of the server SSL certificate. If setting this parameter, also configure
lua_ssl_trusted_certificateinkong.confto specify the CA (or server) certificate used by your Redis server. You may also need to configurelua_ssl_verify_depthaccordingly. Default: true - threshold number
- the default similarity threshold for accepting semantic search results (float). Higher threshold means more results are considered similar.
- username string
- Username to use for Redis connections. If undefined, ACL authentication won't be performed. This requires Redis v6.0.0+. To be compatible with Redis v5.x.y, you can set it to
default. This field is referenceable.
- cloud
Authentication AiGateway Model Api Config Balancer Semantic Vectordb Redis Cloud Authentication - Auth related config for connecting to a Cloud Provider's Redis instance.
- cluster
Ai
Gateway Model Api Config Balancer Semantic Vectordb Redis Cluster - Cluster configuration for the Redis connection.
- connect
Timeout Double - An integer representing a timeout in milliseconds. Must be between 0 and 2^31-2. Default: 2000
- connection
Is BooleanProxied - If the connection to Redis is proxied (e.g. Envoy), set it
true. Set thehostandportto point to the proxy address. Default: false - database Double
- Database to use for the Redis connection when using the
redisstrategy. Default: 0 - dimensions Double
- the desired dimensionality for the vectors. Not Null
- distance
Metric String - the distance metric to use for vector searches. possible known values include one of ["cosine", "euclidean"]; Not Null
- host String
- A string representing a host name, such as example.com. This field is referenceable. Default: "127.0.0.1"
- keepalive
Ai
Gateway Model Api Config Balancer Semantic Vectordb Redis Keepalive - Keepalive configuration for the Redis connection.
- password String
- Password to use for Redis connections. If undefined, no AUTH commands are sent to Redis. This field is referenceable.
- port String
- An integer representing a port number between 0 and 65535, inclusive. This field is referenceable. Default: "6379"
- read
Timeout Double - An integer representing a timeout in milliseconds. Must be between 0 and 2^31-2. Default: 2000
- send
Timeout Double - An integer representing a timeout in milliseconds. Must be between 0 and 2^31-2. Default: 2000
- sentinel
Ai
Gateway Model Api Config Balancer Semantic Vectordb Redis Sentinel - Configuration for Redis Sentinel.
- server
Name String - A string representing an SNI (server name indication) value for TLS. This field is referenceable.
- ssl Boolean
- If set to true, uses SSL to connect to Redis. Default: true
- ssl
Verify Boolean - If set to true, verifies the validity of the server SSL certificate. If setting this parameter, also configure
lua_ssl_trusted_certificateinkong.confto specify the CA (or server) certificate used by your Redis server. You may also need to configurelua_ssl_verify_depthaccordingly. Default: true - threshold Double
- the default similarity threshold for accepting semantic search results (float). Higher threshold means more results are considered similar.
- username String
- Username to use for Redis connections. If undefined, ACL authentication won't be performed. This requires Redis v6.0.0+. To be compatible with Redis v5.x.y, you can set it to
default. This field is referenceable.
- cloud
Authentication AiGateway Model Api Config Balancer Semantic Vectordb Redis Cloud Authentication - Auth related config for connecting to a Cloud Provider's Redis instance.
- cluster
Ai
Gateway Model Api Config Balancer Semantic Vectordb Redis Cluster - Cluster configuration for the Redis connection.
- connect
Timeout number - An integer representing a timeout in milliseconds. Must be between 0 and 2^31-2. Default: 2000
- connection
Is booleanProxied - If the connection to Redis is proxied (e.g. Envoy), set it
true. Set thehostandportto point to the proxy address. Default: false - database number
- Database to use for the Redis connection when using the
redisstrategy. Default: 0 - dimensions number
- the desired dimensionality for the vectors. Not Null
- distance
Metric string - the distance metric to use for vector searches. possible known values include one of ["cosine", "euclidean"]; Not Null
- host string
- A string representing a host name, such as example.com. This field is referenceable. Default: "127.0.0.1"
- keepalive
Ai
Gateway Model Api Config Balancer Semantic Vectordb Redis Keepalive - Keepalive configuration for the Redis connection.
- password string
- Password to use for Redis connections. If undefined, no AUTH commands are sent to Redis. This field is referenceable.
- port string
- An integer representing a port number between 0 and 65535, inclusive. This field is referenceable. Default: "6379"
- read
Timeout number - An integer representing a timeout in milliseconds. Must be between 0 and 2^31-2. Default: 2000
- send
Timeout number - An integer representing a timeout in milliseconds. Must be between 0 and 2^31-2. Default: 2000
- sentinel
Ai
Gateway Model Api Config Balancer Semantic Vectordb Redis Sentinel - Configuration for Redis Sentinel.
- server
Name string - A string representing an SNI (server name indication) value for TLS. This field is referenceable.
- ssl boolean
- If set to true, uses SSL to connect to Redis. Default: true
- ssl
Verify boolean - If set to true, verifies the validity of the server SSL certificate. If setting this parameter, also configure
lua_ssl_trusted_certificateinkong.confto specify the CA (or server) certificate used by your Redis server. You may also need to configurelua_ssl_verify_depthaccordingly. Default: true - threshold number
- the default similarity threshold for accepting semantic search results (float). Higher threshold means more results are considered similar.
- username string
- Username to use for Redis connections. If undefined, ACL authentication won't be performed. This requires Redis v6.0.0+. To be compatible with Redis v5.x.y, you can set it to
default. This field is referenceable.
- cloud_
authentication AiGateway Model Api Config Balancer Semantic Vectordb Redis Cloud Authentication - Auth related config for connecting to a Cloud Provider's Redis instance.
- cluster
Ai
Gateway Model Api Config Balancer Semantic Vectordb Redis Cluster - Cluster configuration for the Redis connection.
- connect_
timeout float - An integer representing a timeout in milliseconds. Must be between 0 and 2^31-2. Default: 2000
- connection_
is_ boolproxied - If the connection to Redis is proxied (e.g. Envoy), set it
true. Set thehostandportto point to the proxy address. Default: false - database float
- Database to use for the Redis connection when using the
redisstrategy. Default: 0 - dimensions float
- the desired dimensionality for the vectors. Not Null
- distance_
metric str - the distance metric to use for vector searches. possible known values include one of ["cosine", "euclidean"]; Not Null
- host str
- A string representing a host name, such as example.com. This field is referenceable. Default: "127.0.0.1"
- keepalive
Ai
Gateway Model Api Config Balancer Semantic Vectordb Redis Keepalive - Keepalive configuration for the Redis connection.
- password str
- Password to use for Redis connections. If undefined, no AUTH commands are sent to Redis. This field is referenceable.
- port str
- An integer representing a port number between 0 and 65535, inclusive. This field is referenceable. Default: "6379"
- read_
timeout float - An integer representing a timeout in milliseconds. Must be between 0 and 2^31-2. Default: 2000
- send_
timeout float - An integer representing a timeout in milliseconds. Must be between 0 and 2^31-2. Default: 2000
- sentinel
Ai
Gateway Model Api Config Balancer Semantic Vectordb Redis Sentinel - Configuration for Redis Sentinel.
- server_
name str - A string representing an SNI (server name indication) value for TLS. This field is referenceable.
- ssl bool
- If set to true, uses SSL to connect to Redis. Default: true
- ssl_
verify bool - If set to true, verifies the validity of the server SSL certificate. If setting this parameter, also configure
lua_ssl_trusted_certificateinkong.confto specify the CA (or server) certificate used by your Redis server. You may also need to configurelua_ssl_verify_depthaccordingly. Default: true - threshold float
- the default similarity threshold for accepting semantic search results (float). Higher threshold means more results are considered similar.
- username str
- Username to use for Redis connections. If undefined, ACL authentication won't be performed. This requires Redis v6.0.0+. To be compatible with Redis v5.x.y, you can set it to
default. This field is referenceable.
- cloud
Authentication Property Map - Auth related config for connecting to a Cloud Provider's Redis instance.
- cluster Property Map
- Cluster configuration for the Redis connection.
- connect
Timeout Number - An integer representing a timeout in milliseconds. Must be between 0 and 2^31-2. Default: 2000
- connection
Is BooleanProxied - If the connection to Redis is proxied (e.g. Envoy), set it
true. Set thehostandportto point to the proxy address. Default: false - database Number
- Database to use for the Redis connection when using the
redisstrategy. Default: 0 - dimensions Number
- the desired dimensionality for the vectors. Not Null
- distance
Metric String - the distance metric to use for vector searches. possible known values include one of ["cosine", "euclidean"]; Not Null
- host String
- A string representing a host name, such as example.com. This field is referenceable. Default: "127.0.0.1"
- keepalive Property Map
- Keepalive configuration for the Redis connection.
- password String
- Password to use for Redis connections. If undefined, no AUTH commands are sent to Redis. This field is referenceable.
- port String
- An integer representing a port number between 0 and 65535, inclusive. This field is referenceable. Default: "6379"
- read
Timeout Number - An integer representing a timeout in milliseconds. Must be between 0 and 2^31-2. Default: 2000
- send
Timeout Number - An integer representing a timeout in milliseconds. Must be between 0 and 2^31-2. Default: 2000
- sentinel Property Map
- Configuration for Redis Sentinel.
- server
Name String - A string representing an SNI (server name indication) value for TLS. This field is referenceable.
- ssl Boolean
- If set to true, uses SSL to connect to Redis. Default: true
- ssl
Verify Boolean - If set to true, verifies the validity of the server SSL certificate. If setting this parameter, also configure
lua_ssl_trusted_certificateinkong.confto specify the CA (or server) certificate used by your Redis server. You may also need to configurelua_ssl_verify_depthaccordingly. Default: true - threshold Number
- the default similarity threshold for accepting semantic search results (float). Higher threshold means more results are considered similar.
- username String
- Username to use for Redis connections. If undefined, ACL authentication won't be performed. This requires Redis v6.0.0+. To be compatible with Redis v5.x.y, you can set it to
default. This field is referenceable.
AiGatewayModelApiConfigBalancerSemanticVectordbRedisCloudAuthentication, AiGatewayModelApiConfigBalancerSemanticVectordbRedisCloudAuthenticationArgs
- Aws
Ai
Gateway Model Api Config Balancer Semantic Vectordb Redis Cloud Authentication Aws - AWS specific configs for connecting to a Cloud Provider's redis instance.
- Azure
Ai
Gateway Model Api Config Balancer Semantic Vectordb Redis Cloud Authentication Azure - Azure specific configs for connecting to a Cloud Provider's redis instance.
- Gcp
Ai
Gateway Model Api Config Balancer Semantic Vectordb Redis Cloud Authentication Gcp - GCP specific configs for connecting to a Cloud Provider's redis instance.
- Aws
Ai
Gateway Model Api Config Balancer Semantic Vectordb Redis Cloud Authentication Aws - AWS specific configs for connecting to a Cloud Provider's redis instance.
- Azure
Ai
Gateway Model Api Config Balancer Semantic Vectordb Redis Cloud Authentication Azure - Azure specific configs for connecting to a Cloud Provider's redis instance.
- Gcp
Ai
Gateway Model Api Config Balancer Semantic Vectordb Redis Cloud Authentication Gcp - GCP specific configs for connecting to a Cloud Provider's redis instance.
- aws
Ai
Gateway Model Api Config Balancer Semantic Vectordb Redis Cloud Authentication Aws - AWS specific configs for connecting to a Cloud Provider's redis instance.
- azure
Ai
Gateway Model Api Config Balancer Semantic Vectordb Redis Cloud Authentication Azure - Azure specific configs for connecting to a Cloud Provider's redis instance.
- gcp
Ai
Gateway Model Api Config Balancer Semantic Vectordb Redis Cloud Authentication Gcp - GCP specific configs for connecting to a Cloud Provider's redis instance.
- aws
Ai
Gateway Model Api Config Balancer Semantic Vectordb Redis Cloud Authentication Aws - AWS specific configs for connecting to a Cloud Provider's redis instance.
- azure
Ai
Gateway Model Api Config Balancer Semantic Vectordb Redis Cloud Authentication Azure - Azure specific configs for connecting to a Cloud Provider's redis instance.
- gcp
Ai
Gateway Model Api Config Balancer Semantic Vectordb Redis Cloud Authentication Gcp - GCP specific configs for connecting to a Cloud Provider's redis instance.
- aws
Ai
Gateway Model Api Config Balancer Semantic Vectordb Redis Cloud Authentication Aws - AWS specific configs for connecting to a Cloud Provider's redis instance.
- azure
Ai
Gateway Model Api Config Balancer Semantic Vectordb Redis Cloud Authentication Azure - Azure specific configs for connecting to a Cloud Provider's redis instance.
- gcp
Ai
Gateway Model Api Config Balancer Semantic Vectordb Redis Cloud Authentication Gcp - GCP specific configs for connecting to a Cloud Provider's redis instance.
- aws Property Map
- AWS specific configs for connecting to a Cloud Provider's redis instance.
- azure Property Map
- Azure specific configs for connecting to a Cloud Provider's redis instance.
- gcp Property Map
- GCP specific configs for connecting to a Cloud Provider's redis instance.
AiGatewayModelApiConfigBalancerSemanticVectordbRedisCloudAuthenticationAws, AiGatewayModelApiConfigBalancerSemanticVectordbRedisCloudAuthenticationAwsArgs
- Access
Key stringId - AWS Access Key ID to be used for authentication. This field is referenceable.
- Assume
Role stringArn - The ARN of the IAM role to assume for generating ElastiCache IAM authentication tokens. This field is referenceable.
- Cache
Name string - The name of the AWS Elasticache cluster. This field is referenceable.
- Is
Serverless bool - This flag specifies whether the cluster is serverless. Default: true
- Region string
- The region of the AWS ElastiCache cluster. This field is referenceable.
- Role
Session stringName - The session name for the temporary credentials when assuming the IAM role. This field is referenceable.
- Secret
Access stringKey - AWS Secret Access Key. This field is referenceable.
- Access
Key stringId - AWS Access Key ID to be used for authentication. This field is referenceable.
- Assume
Role stringArn - The ARN of the IAM role to assume for generating ElastiCache IAM authentication tokens. This field is referenceable.
- Cache
Name string - The name of the AWS Elasticache cluster. This field is referenceable.
- Is
Serverless bool - This flag specifies whether the cluster is serverless. Default: true
- Region string
- The region of the AWS ElastiCache cluster. This field is referenceable.
- Role
Session stringName - The session name for the temporary credentials when assuming the IAM role. This field is referenceable.
- Secret
Access stringKey - AWS Secret Access Key. This field is referenceable.
- access_
key_ stringid - AWS Access Key ID to be used for authentication. This field is referenceable.
- assume_
role_ stringarn - The ARN of the IAM role to assume for generating ElastiCache IAM authentication tokens. This field is referenceable.
- cache_
name string - The name of the AWS Elasticache cluster. This field is referenceable.
- is_
serverless bool - This flag specifies whether the cluster is serverless. Default: true
- region string
- The region of the AWS ElastiCache cluster. This field is referenceable.
- role_
session_ stringname - The session name for the temporary credentials when assuming the IAM role. This field is referenceable.
- secret_
access_ stringkey - AWS Secret Access Key. This field is referenceable.
- access
Key StringId - AWS Access Key ID to be used for authentication. This field is referenceable.
- assume
Role StringArn - The ARN of the IAM role to assume for generating ElastiCache IAM authentication tokens. This field is referenceable.
- cache
Name String - The name of the AWS Elasticache cluster. This field is referenceable.
- is
Serverless Boolean - This flag specifies whether the cluster is serverless. Default: true
- region String
- The region of the AWS ElastiCache cluster. This field is referenceable.
- role
Session StringName - The session name for the temporary credentials when assuming the IAM role. This field is referenceable.
- secret
Access StringKey - AWS Secret Access Key. This field is referenceable.
- access
Key stringId - AWS Access Key ID to be used for authentication. This field is referenceable.
- assume
Role stringArn - The ARN of the IAM role to assume for generating ElastiCache IAM authentication tokens. This field is referenceable.
- cache
Name string - The name of the AWS Elasticache cluster. This field is referenceable.
- is
Serverless boolean - This flag specifies whether the cluster is serverless. Default: true
- region string
- The region of the AWS ElastiCache cluster. This field is referenceable.
- role
Session stringName - The session name for the temporary credentials when assuming the IAM role. This field is referenceable.
- secret
Access stringKey - AWS Secret Access Key. This field is referenceable.
- access_
key_ strid - AWS Access Key ID to be used for authentication. This field is referenceable.
- assume_
role_ strarn - The ARN of the IAM role to assume for generating ElastiCache IAM authentication tokens. This field is referenceable.
- cache_
name str - The name of the AWS Elasticache cluster. This field is referenceable.
- is_
serverless bool - This flag specifies whether the cluster is serverless. Default: true
- region str
- The region of the AWS ElastiCache cluster. This field is referenceable.
- role_
session_ strname - The session name for the temporary credentials when assuming the IAM role. This field is referenceable.
- secret_
access_ strkey - AWS Secret Access Key. This field is referenceable.
- access
Key StringId - AWS Access Key ID to be used for authentication. This field is referenceable.
- assume
Role StringArn - The ARN of the IAM role to assume for generating ElastiCache IAM authentication tokens. This field is referenceable.
- cache
Name String - The name of the AWS Elasticache cluster. This field is referenceable.
- is
Serverless Boolean - This flag specifies whether the cluster is serverless. Default: true
- region String
- The region of the AWS ElastiCache cluster. This field is referenceable.
- role
Session StringName - The session name for the temporary credentials when assuming the IAM role. This field is referenceable.
- secret
Access StringKey - AWS Secret Access Key. This field is referenceable.
AiGatewayModelApiConfigBalancerSemanticVectordbRedisCloudAuthenticationAzure, AiGatewayModelApiConfigBalancerSemanticVectordbRedisCloudAuthenticationAzureArgs
- Client
Id string - Azure Client ID. This field is referenceable.
- Client
Secret string - Azure Client Secret. This field is referenceable.
- Tenant
Id string - Azure Tenant ID. This field is referenceable.
- Client
Id string - Azure Client ID. This field is referenceable.
- Client
Secret string - Azure Client Secret. This field is referenceable.
- Tenant
Id string - Azure Tenant ID. This field is referenceable.
- client_
id string - Azure Client ID. This field is referenceable.
- client_
secret string - Azure Client Secret. This field is referenceable.
- tenant_
id string - Azure Tenant ID. This field is referenceable.
- client
Id String - Azure Client ID. This field is referenceable.
- client
Secret String - Azure Client Secret. This field is referenceable.
- tenant
Id String - Azure Tenant ID. This field is referenceable.
- client
Id string - Azure Client ID. This field is referenceable.
- client
Secret string - Azure Client Secret. This field is referenceable.
- tenant
Id string - Azure Tenant ID. This field is referenceable.
- client_
id str - Azure Client ID. This field is referenceable.
- client_
secret str - Azure Client Secret. This field is referenceable.
- tenant_
id str - Azure Tenant ID. This field is referenceable.
- client
Id String - Azure Client ID. This field is referenceable.
- client
Secret String - Azure Client Secret. This field is referenceable.
- tenant
Id String - Azure Tenant ID. This field is referenceable.
AiGatewayModelApiConfigBalancerSemanticVectordbRedisCloudAuthenticationGcp, AiGatewayModelApiConfigBalancerSemanticVectordbRedisCloudAuthenticationGcpArgs
- Service
Account stringJson - GCP Service Account JSON. This field is referenceable.
- Service
Account stringJson - GCP Service Account JSON. This field is referenceable.
- service_
account_ stringjson - GCP Service Account JSON. This field is referenceable.
- service
Account StringJson - GCP Service Account JSON. This field is referenceable.
- service
Account stringJson - GCP Service Account JSON. This field is referenceable.
- service_
account_ strjson - GCP Service Account JSON. This field is referenceable.
- service
Account StringJson - GCP Service Account JSON. This field is referenceable.
AiGatewayModelApiConfigBalancerSemanticVectordbRedisCluster, AiGatewayModelApiConfigBalancerSemanticVectordbRedisClusterArgs
- Max
Redirections double - Maximum retry attempts for redirection. Default: 5
- Nodes
List<Ai
Gateway Model Api Config Balancer Semantic Vectordb Redis Cluster Node> - Cluster addresses to use for Redis connections when the
redisstrategy is defined. Defining this field implies using a Redis Cluster. The minimum length of the array is 1 element.
- Max
Redirections float64 - Maximum retry attempts for redirection. Default: 5
- Nodes
[]Ai
Gateway Model Api Config Balancer Semantic Vectordb Redis Cluster Node - Cluster addresses to use for Redis connections when the
redisstrategy is defined. Defining this field implies using a Redis Cluster. The minimum length of the array is 1 element.
- max_
redirections number - Maximum retry attempts for redirection. Default: 5
- nodes list(object)
- Cluster addresses to use for Redis connections when the
redisstrategy is defined. Defining this field implies using a Redis Cluster. The minimum length of the array is 1 element.
- max
Redirections Double - Maximum retry attempts for redirection. Default: 5
- nodes
List<Ai
Gateway Model Api Config Balancer Semantic Vectordb Redis Cluster Node> - Cluster addresses to use for Redis connections when the
redisstrategy is defined. Defining this field implies using a Redis Cluster. The minimum length of the array is 1 element.
- max
Redirections number - Maximum retry attempts for redirection. Default: 5
- nodes
Ai
Gateway Model Api Config Balancer Semantic Vectordb Redis Cluster Node[] - Cluster addresses to use for Redis connections when the
redisstrategy is defined. Defining this field implies using a Redis Cluster. The minimum length of the array is 1 element.
- max_
redirections float - Maximum retry attempts for redirection. Default: 5
- nodes
Sequence[Ai
Gateway Model Api Config Balancer Semantic Vectordb Redis Cluster Node] - Cluster addresses to use for Redis connections when the
redisstrategy is defined. Defining this field implies using a Redis Cluster. The minimum length of the array is 1 element.
- max
Redirections Number - Maximum retry attempts for redirection. Default: 5
- nodes List<Property Map>
- Cluster addresses to use for Redis connections when the
redisstrategy is defined. Defining this field implies using a Redis Cluster. The minimum length of the array is 1 element.
AiGatewayModelApiConfigBalancerSemanticVectordbRedisClusterNode, AiGatewayModelApiConfigBalancerSemanticVectordbRedisClusterNodeArgs
AiGatewayModelApiConfigBalancerSemanticVectordbRedisKeepalive, AiGatewayModelApiConfigBalancerSemanticVectordbRedisKeepaliveArgs
- Backlog double
- Limits the total number of opened connections for a pool. If the connection pool is full, connection queues above the limit go into the backlog queue. If the backlog queue is full, subsequent connect operations fail and return
nil. Queued operations (subject to set timeouts) resume once the number of connections in the pool is less thanpool_size. If latency is high or throughput is low, try increasing this value. Empirically, this value is larger thanpool_size. - Pool
Size double - The size limit for every cosocket connection pool associated with every remote server, per worker process. If neither
pool_sizenorbacklogis specified, no pool is created. Ifpool_sizeisn't specified butbacklogis specified, then the pool uses the default value. Try to increase (e.g. 512) this value if latency is high or throughput is low. Default: 256
- Backlog float64
- Limits the total number of opened connections for a pool. If the connection pool is full, connection queues above the limit go into the backlog queue. If the backlog queue is full, subsequent connect operations fail and return
nil. Queued operations (subject to set timeouts) resume once the number of connections in the pool is less thanpool_size. If latency is high or throughput is low, try increasing this value. Empirically, this value is larger thanpool_size. - Pool
Size float64 - The size limit for every cosocket connection pool associated with every remote server, per worker process. If neither
pool_sizenorbacklogis specified, no pool is created. Ifpool_sizeisn't specified butbacklogis specified, then the pool uses the default value. Try to increase (e.g. 512) this value if latency is high or throughput is low. Default: 256
- backlog number
- Limits the total number of opened connections for a pool. If the connection pool is full, connection queues above the limit go into the backlog queue. If the backlog queue is full, subsequent connect operations fail and return
nil. Queued operations (subject to set timeouts) resume once the number of connections in the pool is less thanpool_size. If latency is high or throughput is low, try increasing this value. Empirically, this value is larger thanpool_size. - pool_
size number - The size limit for every cosocket connection pool associated with every remote server, per worker process. If neither
pool_sizenorbacklogis specified, no pool is created. Ifpool_sizeisn't specified butbacklogis specified, then the pool uses the default value. Try to increase (e.g. 512) this value if latency is high or throughput is low. Default: 256
- backlog Double
- Limits the total number of opened connections for a pool. If the connection pool is full, connection queues above the limit go into the backlog queue. If the backlog queue is full, subsequent connect operations fail and return
nil. Queued operations (subject to set timeouts) resume once the number of connections in the pool is less thanpool_size. If latency is high or throughput is low, try increasing this value. Empirically, this value is larger thanpool_size. - pool
Size Double - The size limit for every cosocket connection pool associated with every remote server, per worker process. If neither
pool_sizenorbacklogis specified, no pool is created. Ifpool_sizeisn't specified butbacklogis specified, then the pool uses the default value. Try to increase (e.g. 512) this value if latency is high or throughput is low. Default: 256
- backlog number
- Limits the total number of opened connections for a pool. If the connection pool is full, connection queues above the limit go into the backlog queue. If the backlog queue is full, subsequent connect operations fail and return
nil. Queued operations (subject to set timeouts) resume once the number of connections in the pool is less thanpool_size. If latency is high or throughput is low, try increasing this value. Empirically, this value is larger thanpool_size. - pool
Size number - The size limit for every cosocket connection pool associated with every remote server, per worker process. If neither
pool_sizenorbacklogis specified, no pool is created. Ifpool_sizeisn't specified butbacklogis specified, then the pool uses the default value. Try to increase (e.g. 512) this value if latency is high or throughput is low. Default: 256
- backlog float
- Limits the total number of opened connections for a pool. If the connection pool is full, connection queues above the limit go into the backlog queue. If the backlog queue is full, subsequent connect operations fail and return
nil. Queued operations (subject to set timeouts) resume once the number of connections in the pool is less thanpool_size. If latency is high or throughput is low, try increasing this value. Empirically, this value is larger thanpool_size. - pool_
size float - The size limit for every cosocket connection pool associated with every remote server, per worker process. If neither
pool_sizenorbacklogis specified, no pool is created. Ifpool_sizeisn't specified butbacklogis specified, then the pool uses the default value. Try to increase (e.g. 512) this value if latency is high or throughput is low. Default: 256
- backlog Number
- Limits the total number of opened connections for a pool. If the connection pool is full, connection queues above the limit go into the backlog queue. If the backlog queue is full, subsequent connect operations fail and return
nil. Queued operations (subject to set timeouts) resume once the number of connections in the pool is less thanpool_size. If latency is high or throughput is low, try increasing this value. Empirically, this value is larger thanpool_size. - pool
Size Number - The size limit for every cosocket connection pool associated with every remote server, per worker process. If neither
pool_sizenorbacklogis specified, no pool is created. Ifpool_sizeisn't specified butbacklogis specified, then the pool uses the default value. Try to increase (e.g. 512) this value if latency is high or throughput is low. Default: 256
AiGatewayModelApiConfigBalancerSemanticVectordbRedisSentinel, AiGatewayModelApiConfigBalancerSemanticVectordbRedisSentinelArgs
- Master string
- Sentinel master to use for Redis connections. Defining this value implies using Redis Sentinel.
- Nodes
List<Ai
Gateway Model Api Config Balancer Semantic Vectordb Redis Sentinel Node> - Sentinel node addresses to use for Redis connections when the
redisstrategy is defined. Defining this field implies using a Redis Sentinel. The minimum length of the array is 1 element. - Password string
- Sentinel password to authenticate with a Redis Sentinel instance. If undefined, no AUTH commands are sent to Redis Sentinels. This field is referenceable.
- Role string
- Sentinel role to use for Redis connections when the
redisstrategy is defined. Defining this value implies using Redis Sentinel. possible known values include one of ["any", "master", "slave"] - Username string
- Sentinel username to authenticate with a Redis Sentinel instance. If undefined, ACL authentication won't be performed. This requires Redis v6.2.0+. This field is referenceable.
- Master string
- Sentinel master to use for Redis connections. Defining this value implies using Redis Sentinel.
- Nodes
[]Ai
Gateway Model Api Config Balancer Semantic Vectordb Redis Sentinel Node - Sentinel node addresses to use for Redis connections when the
redisstrategy is defined. Defining this field implies using a Redis Sentinel. The minimum length of the array is 1 element. - Password string
- Sentinel password to authenticate with a Redis Sentinel instance. If undefined, no AUTH commands are sent to Redis Sentinels. This field is referenceable.
- Role string
- Sentinel role to use for Redis connections when the
redisstrategy is defined. Defining this value implies using Redis Sentinel. possible known values include one of ["any", "master", "slave"] - Username string
- Sentinel username to authenticate with a Redis Sentinel instance. If undefined, ACL authentication won't be performed. This requires Redis v6.2.0+. This field is referenceable.
- master string
- Sentinel master to use for Redis connections. Defining this value implies using Redis Sentinel.
- nodes list(object)
- Sentinel node addresses to use for Redis connections when the
redisstrategy is defined. Defining this field implies using a Redis Sentinel. The minimum length of the array is 1 element. - password string
- Sentinel password to authenticate with a Redis Sentinel instance. If undefined, no AUTH commands are sent to Redis Sentinels. This field is referenceable.
- role string
- Sentinel role to use for Redis connections when the
redisstrategy is defined. Defining this value implies using Redis Sentinel. possible known values include one of ["any", "master", "slave"] - username string
- Sentinel username to authenticate with a Redis Sentinel instance. If undefined, ACL authentication won't be performed. This requires Redis v6.2.0+. This field is referenceable.
- master String
- Sentinel master to use for Redis connections. Defining this value implies using Redis Sentinel.
- nodes
List<Ai
Gateway Model Api Config Balancer Semantic Vectordb Redis Sentinel Node> - Sentinel node addresses to use for Redis connections when the
redisstrategy is defined. Defining this field implies using a Redis Sentinel. The minimum length of the array is 1 element. - password String
- Sentinel password to authenticate with a Redis Sentinel instance. If undefined, no AUTH commands are sent to Redis Sentinels. This field is referenceable.
- role String
- Sentinel role to use for Redis connections when the
redisstrategy is defined. Defining this value implies using Redis Sentinel. possible known values include one of ["any", "master", "slave"] - username String
- Sentinel username to authenticate with a Redis Sentinel instance. If undefined, ACL authentication won't be performed. This requires Redis v6.2.0+. This field is referenceable.
- master string
- Sentinel master to use for Redis connections. Defining this value implies using Redis Sentinel.
- nodes
Ai
Gateway Model Api Config Balancer Semantic Vectordb Redis Sentinel Node[] - Sentinel node addresses to use for Redis connections when the
redisstrategy is defined. Defining this field implies using a Redis Sentinel. The minimum length of the array is 1 element. - password string
- Sentinel password to authenticate with a Redis Sentinel instance. If undefined, no AUTH commands are sent to Redis Sentinels. This field is referenceable.
- role string
- Sentinel role to use for Redis connections when the
redisstrategy is defined. Defining this value implies using Redis Sentinel. possible known values include one of ["any", "master", "slave"] - username string
- Sentinel username to authenticate with a Redis Sentinel instance. If undefined, ACL authentication won't be performed. This requires Redis v6.2.0+. This field is referenceable.
- master str
- Sentinel master to use for Redis connections. Defining this value implies using Redis Sentinel.
- nodes
Sequence[Ai
Gateway Model Api Config Balancer Semantic Vectordb Redis Sentinel Node] - Sentinel node addresses to use for Redis connections when the
redisstrategy is defined. Defining this field implies using a Redis Sentinel. The minimum length of the array is 1 element. - password str
- Sentinel password to authenticate with a Redis Sentinel instance. If undefined, no AUTH commands are sent to Redis Sentinels. This field is referenceable.
- role str
- Sentinel role to use for Redis connections when the
redisstrategy is defined. Defining this value implies using Redis Sentinel. possible known values include one of ["any", "master", "slave"] - username str
- Sentinel username to authenticate with a Redis Sentinel instance. If undefined, ACL authentication won't be performed. This requires Redis v6.2.0+. This field is referenceable.
- master String
- Sentinel master to use for Redis connections. Defining this value implies using Redis Sentinel.
- nodes List<Property Map>
- Sentinel node addresses to use for Redis connections when the
redisstrategy is defined. Defining this field implies using a Redis Sentinel. The minimum length of the array is 1 element. - password String
- Sentinel password to authenticate with a Redis Sentinel instance. If undefined, no AUTH commands are sent to Redis Sentinels. This field is referenceable.
- role String
- Sentinel role to use for Redis connections when the
redisstrategy is defined. Defining this value implies using Redis Sentinel. possible known values include one of ["any", "master", "slave"] - username String
- Sentinel username to authenticate with a Redis Sentinel instance. If undefined, ACL authentication won't be performed. This requires Redis v6.2.0+. This field is referenceable.
AiGatewayModelApiConfigBalancerSemanticVectordbRedisSentinelNode, AiGatewayModelApiConfigBalancerSemanticVectordbRedisSentinelNodeArgs
AiGatewayModelApiConfigLogging, AiGatewayModelApiConfigLoggingArgs
- Payloads bool
- Default: false
- Payloads bool
- Default: false
- payloads bool
- Default: false
- payloads Boolean
- Default: false
- payloads boolean
- Default: false
- payloads bool
- Default: false
- payloads Boolean
- Default: false
AiGatewayModelApiConfigProxy, AiGatewayModelApiConfigProxyArgs
- Auth
Ai
Gateway Model Api Config Proxy Auth - Credentials used to authenticate to the proxy server.
- Http
Proxy AiGateway Model Api Config Proxy Http Proxy - HTTP proxy server to route plaintext outbound requests through.
- Https
Proxy AiGateway Model Api Config Proxy Https Proxy - HTTPS proxy server to route TLS outbound requests through.
- No
Proxy string - Comma-separated list of hosts that should not be proxied.
- Proxy
Scheme string - The proxy scheme to use when connecting to the proxy server. Default: "http"; must be "http"
- Auth
Ai
Gateway Model Api Config Proxy Auth - Credentials used to authenticate to the proxy server.
- Http
Proxy AiGateway Model Api Config Proxy Http Proxy - HTTP proxy server to route plaintext outbound requests through.
- Https
Proxy AiGateway Model Api Config Proxy Https Proxy - HTTPS proxy server to route TLS outbound requests through.
- No
Proxy string - Comma-separated list of hosts that should not be proxied.
- Proxy
Scheme string - The proxy scheme to use when connecting to the proxy server. Default: "http"; must be "http"
- auth object
- Credentials used to authenticate to the proxy server.
- http_
proxy object - HTTP proxy server to route plaintext outbound requests through.
- https_
proxy object - HTTPS proxy server to route TLS outbound requests through.
- no_
proxy string - Comma-separated list of hosts that should not be proxied.
- proxy_
scheme string - The proxy scheme to use when connecting to the proxy server. Default: "http"; must be "http"
- auth
Ai
Gateway Model Api Config Proxy Auth - Credentials used to authenticate to the proxy server.
- http
Proxy AiGateway Model Api Config Proxy Http Proxy - HTTP proxy server to route plaintext outbound requests through.
- https
Proxy AiGateway Model Api Config Proxy Https Proxy - HTTPS proxy server to route TLS outbound requests through.
- no
Proxy String - Comma-separated list of hosts that should not be proxied.
- proxy
Scheme String - The proxy scheme to use when connecting to the proxy server. Default: "http"; must be "http"
- auth
Ai
Gateway Model Api Config Proxy Auth - Credentials used to authenticate to the proxy server.
- http
Proxy AiGateway Model Api Config Proxy Http Proxy - HTTP proxy server to route plaintext outbound requests through.
- https
Proxy AiGateway Model Api Config Proxy Https Proxy - HTTPS proxy server to route TLS outbound requests through.
- no
Proxy string - Comma-separated list of hosts that should not be proxied.
- proxy
Scheme string - The proxy scheme to use when connecting to the proxy server. Default: "http"; must be "http"
- auth
Ai
Gateway Model Api Config Proxy Auth - Credentials used to authenticate to the proxy server.
- http_
proxy AiGateway Model Api Config Proxy Http Proxy - HTTP proxy server to route plaintext outbound requests through.
- https_
proxy AiGateway Model Api Config Proxy Https Proxy - HTTPS proxy server to route TLS outbound requests through.
- no_
proxy str - Comma-separated list of hosts that should not be proxied.
- proxy_
scheme str - The proxy scheme to use when connecting to the proxy server. Default: "http"; must be "http"
- auth Property Map
- Credentials used to authenticate to the proxy server.
- http
Proxy Property Map - HTTP proxy server to route plaintext outbound requests through.
- https
Proxy Property Map - HTTPS proxy server to route TLS outbound requests through.
- no
Proxy String - Comma-separated list of hosts that should not be proxied.
- proxy
Scheme String - The proxy scheme to use when connecting to the proxy server. Default: "http"; must be "http"
AiGatewayModelApiConfigProxyAuth, AiGatewayModelApiConfigProxyAuthArgs
- Password string
- The password to use for proxy authentication. This field is referenceable.
- Username string
- The username to use for proxy authentication. This field is referenceable.
- Password string
- The password to use for proxy authentication. This field is referenceable.
- Username string
- The username to use for proxy authentication. This field is referenceable.
- password string
- The password to use for proxy authentication. This field is referenceable.
- username string
- The username to use for proxy authentication. This field is referenceable.
- password String
- The password to use for proxy authentication. This field is referenceable.
- username String
- The username to use for proxy authentication. This field is referenceable.
- password string
- The password to use for proxy authentication. This field is referenceable.
- username string
- The username to use for proxy authentication. This field is referenceable.
- password str
- The password to use for proxy authentication. This field is referenceable.
- username str
- The username to use for proxy authentication. This field is referenceable.
- password String
- The password to use for proxy authentication. This field is referenceable.
- username String
- The username to use for proxy authentication. This field is referenceable.
AiGatewayModelApiConfigProxyHttpProxy, AiGatewayModelApiConfigProxyHttpProxyArgs
AiGatewayModelApiConfigProxyHttpsProxy, AiGatewayModelApiConfigProxyHttpsProxyArgs
AiGatewayModelApiConfigRoute, AiGatewayModelApiConfigRouteArgs
- Headers Dictionary<string, string>
- One or more lists of values indexed by header name that will cause this route to match if present in the request. The
Hostheader cannot be used with this attribute: hosts should be specified using thehostsattribute. Whenheaderscontains only one value and that value starts with the special prefix~*, the value is interpreted as a regular expression. - Hosts List<string>
- A list of domain names that match this route. Note that the hosts value is case sensitive.
- Https
Redirect doubleStatus Code - The status code Kong responds with when all properties of a route match except the protocol i.e. if the protocol of the request is
HTTPinstead ofHTTPS.Locationheader is injected by Kong if the field is set to 301, 302, 307 or 308. Note: This config applies only if the route is configured to only accept thehttpsprotocol. Default: 426 - Methods List<string>
- A list of HTTP methods that match this route.
- Model
Ai
Gateway Model Api Config Route Model - Configuration for overriding routing to this model using a selector. When no selector location is set, the format default selector is used. When values are not set, the model name is used as the selector value.
- Paths List<string>
- A list of paths that match this route.
- Preserve
Host bool - When matching a route via one of the
hostsdomain names, use the requestHostheader in the upstream request headers. If set tofalse, the upstreamHostheader will be that of the service'shost. Default: false - Protocols List<string>
- An array of the protocols this route should allow. See the route Object section for a list of accepted protocols. When set to only
https, HTTP requests are answered with an upgrade error. When set to onlyhttp, HTTPS requests are answered with an error. Default: ["http","https"] - Regex
Priority double - A number used to choose which route resolves a given request when several routes match it using regexes simultaneously. When two routes match the path and have the same
regex_priority, the older one (lowestcreated_at) is used. Note that the priority for non-regex routes is different (longer non-regex routes are matched before shorter ones). Default: 0 - Request
Buffering bool - Whether to enable request body buffering or not. With HTTP 1.1, it may make sense to turn this off on services that receive data with chunked transfer encoding. Default: true
- Response
Buffering bool - Whether to enable response body buffering or not. With HTTP 1.1, it may make sense to turn this off on services that send data with chunked transfer encoding. Default: true
- Strip
Path bool - When matching a route via one of the
paths, strip the matching prefix from the upstream request URL. Default: true - List<string>
- An optional set of strings associated with the route for grouping and filtering.
- Headers map[string]string
- One or more lists of values indexed by header name that will cause this route to match if present in the request. The
Hostheader cannot be used with this attribute: hosts should be specified using thehostsattribute. Whenheaderscontains only one value and that value starts with the special prefix~*, the value is interpreted as a regular expression. - Hosts []string
- A list of domain names that match this route. Note that the hosts value is case sensitive.
- Https
Redirect float64Status Code - The status code Kong responds with when all properties of a route match except the protocol i.e. if the protocol of the request is
HTTPinstead ofHTTPS.Locationheader is injected by Kong if the field is set to 301, 302, 307 or 308. Note: This config applies only if the route is configured to only accept thehttpsprotocol. Default: 426 - Methods []string
- A list of HTTP methods that match this route.
- Model
Ai
Gateway Model Api Config Route Model - Configuration for overriding routing to this model using a selector. When no selector location is set, the format default selector is used. When values are not set, the model name is used as the selector value.
- Paths []string
- A list of paths that match this route.
- Preserve
Host bool - When matching a route via one of the
hostsdomain names, use the requestHostheader in the upstream request headers. If set tofalse, the upstreamHostheader will be that of the service'shost. Default: false - Protocols []string
- An array of the protocols this route should allow. See the route Object section for a list of accepted protocols. When set to only
https, HTTP requests are answered with an upgrade error. When set to onlyhttp, HTTPS requests are answered with an error. Default: ["http","https"] - Regex
Priority float64 - A number used to choose which route resolves a given request when several routes match it using regexes simultaneously. When two routes match the path and have the same
regex_priority, the older one (lowestcreated_at) is used. Note that the priority for non-regex routes is different (longer non-regex routes are matched before shorter ones). Default: 0 - Request
Buffering bool - Whether to enable request body buffering or not. With HTTP 1.1, it may make sense to turn this off on services that receive data with chunked transfer encoding. Default: true
- Response
Buffering bool - Whether to enable response body buffering or not. With HTTP 1.1, it may make sense to turn this off on services that send data with chunked transfer encoding. Default: true
- Strip
Path bool - When matching a route via one of the
paths, strip the matching prefix from the upstream request URL. Default: true - []string
- An optional set of strings associated with the route for grouping and filtering.
- headers map(string)
- One or more lists of values indexed by header name that will cause this route to match if present in the request. The
Hostheader cannot be used with this attribute: hosts should be specified using thehostsattribute. Whenheaderscontains only one value and that value starts with the special prefix~*, the value is interpreted as a regular expression. - hosts list(string)
- A list of domain names that match this route. Note that the hosts value is case sensitive.
- https_
redirect_ numberstatus_ code - The status code Kong responds with when all properties of a route match except the protocol i.e. if the protocol of the request is
HTTPinstead ofHTTPS.Locationheader is injected by Kong if the field is set to 301, 302, 307 or 308. Note: This config applies only if the route is configured to only accept thehttpsprotocol. Default: 426 - methods list(string)
- A list of HTTP methods that match this route.
- model object
- Configuration for overriding routing to this model using a selector. When no selector location is set, the format default selector is used. When values are not set, the model name is used as the selector value.
- paths list(string)
- A list of paths that match this route.
- preserve_
host bool - When matching a route via one of the
hostsdomain names, use the requestHostheader in the upstream request headers. If set tofalse, the upstreamHostheader will be that of the service'shost. Default: false - protocols list(string)
- An array of the protocols this route should allow. See the route Object section for a list of accepted protocols. When set to only
https, HTTP requests are answered with an upgrade error. When set to onlyhttp, HTTPS requests are answered with an error. Default: ["http","https"] - regex_
priority number - A number used to choose which route resolves a given request when several routes match it using regexes simultaneously. When two routes match the path and have the same
regex_priority, the older one (lowestcreated_at) is used. Note that the priority for non-regex routes is different (longer non-regex routes are matched before shorter ones). Default: 0 - request_
buffering bool - Whether to enable request body buffering or not. With HTTP 1.1, it may make sense to turn this off on services that receive data with chunked transfer encoding. Default: true
- response_
buffering bool - Whether to enable response body buffering or not. With HTTP 1.1, it may make sense to turn this off on services that send data with chunked transfer encoding. Default: true
- strip_
path bool - When matching a route via one of the
paths, strip the matching prefix from the upstream request URL. Default: true - list(string)
- An optional set of strings associated with the route for grouping and filtering.
- headers Map<String,String>
- One or more lists of values indexed by header name that will cause this route to match if present in the request. The
Hostheader cannot be used with this attribute: hosts should be specified using thehostsattribute. Whenheaderscontains only one value and that value starts with the special prefix~*, the value is interpreted as a regular expression. - hosts List<String>
- A list of domain names that match this route. Note that the hosts value is case sensitive.
- https
Redirect DoubleStatus Code - The status code Kong responds with when all properties of a route match except the protocol i.e. if the protocol of the request is
HTTPinstead ofHTTPS.Locationheader is injected by Kong if the field is set to 301, 302, 307 or 308. Note: This config applies only if the route is configured to only accept thehttpsprotocol. Default: 426 - methods List<String>
- A list of HTTP methods that match this route.
- model
Ai
Gateway Model Api Config Route Model - Configuration for overriding routing to this model using a selector. When no selector location is set, the format default selector is used. When values are not set, the model name is used as the selector value.
- paths List<String>
- A list of paths that match this route.
- preserve
Host Boolean - When matching a route via one of the
hostsdomain names, use the requestHostheader in the upstream request headers. If set tofalse, the upstreamHostheader will be that of the service'shost. Default: false - protocols List<String>
- An array of the protocols this route should allow. See the route Object section for a list of accepted protocols. When set to only
https, HTTP requests are answered with an upgrade error. When set to onlyhttp, HTTPS requests are answered with an error. Default: ["http","https"] - regex
Priority Double - A number used to choose which route resolves a given request when several routes match it using regexes simultaneously. When two routes match the path and have the same
regex_priority, the older one (lowestcreated_at) is used. Note that the priority for non-regex routes is different (longer non-regex routes are matched before shorter ones). Default: 0 - request
Buffering Boolean - Whether to enable request body buffering or not. With HTTP 1.1, it may make sense to turn this off on services that receive data with chunked transfer encoding. Default: true
- response
Buffering Boolean - Whether to enable response body buffering or not. With HTTP 1.1, it may make sense to turn this off on services that send data with chunked transfer encoding. Default: true
- strip
Path Boolean - When matching a route via one of the
paths, strip the matching prefix from the upstream request URL. Default: true - List<String>
- An optional set of strings associated with the route for grouping and filtering.
- headers {[key: string]: string}
- One or more lists of values indexed by header name that will cause this route to match if present in the request. The
Hostheader cannot be used with this attribute: hosts should be specified using thehostsattribute. Whenheaderscontains only one value and that value starts with the special prefix~*, the value is interpreted as a regular expression. - hosts string[]
- A list of domain names that match this route. Note that the hosts value is case sensitive.
- https
Redirect numberStatus Code - The status code Kong responds with when all properties of a route match except the protocol i.e. if the protocol of the request is
HTTPinstead ofHTTPS.Locationheader is injected by Kong if the field is set to 301, 302, 307 or 308. Note: This config applies only if the route is configured to only accept thehttpsprotocol. Default: 426 - methods string[]
- A list of HTTP methods that match this route.
- model
Ai
Gateway Model Api Config Route Model - Configuration for overriding routing to this model using a selector. When no selector location is set, the format default selector is used. When values are not set, the model name is used as the selector value.
- paths string[]
- A list of paths that match this route.
- preserve
Host boolean - When matching a route via one of the
hostsdomain names, use the requestHostheader in the upstream request headers. If set tofalse, the upstreamHostheader will be that of the service'shost. Default: false - protocols string[]
- An array of the protocols this route should allow. See the route Object section for a list of accepted protocols. When set to only
https, HTTP requests are answered with an upgrade error. When set to onlyhttp, HTTPS requests are answered with an error. Default: ["http","https"] - regex
Priority number - A number used to choose which route resolves a given request when several routes match it using regexes simultaneously. When two routes match the path and have the same
regex_priority, the older one (lowestcreated_at) is used. Note that the priority for non-regex routes is different (longer non-regex routes are matched before shorter ones). Default: 0 - request
Buffering boolean - Whether to enable request body buffering or not. With HTTP 1.1, it may make sense to turn this off on services that receive data with chunked transfer encoding. Default: true
- response
Buffering boolean - Whether to enable response body buffering or not. With HTTP 1.1, it may make sense to turn this off on services that send data with chunked transfer encoding. Default: true
- strip
Path boolean - When matching a route via one of the
paths, strip the matching prefix from the upstream request URL. Default: true - string[]
- An optional set of strings associated with the route for grouping and filtering.
- headers Mapping[str, str]
- One or more lists of values indexed by header name that will cause this route to match if present in the request. The
Hostheader cannot be used with this attribute: hosts should be specified using thehostsattribute. Whenheaderscontains only one value and that value starts with the special prefix~*, the value is interpreted as a regular expression. - hosts Sequence[str]
- A list of domain names that match this route. Note that the hosts value is case sensitive.
- https_
redirect_ floatstatus_ code - The status code Kong responds with when all properties of a route match except the protocol i.e. if the protocol of the request is
HTTPinstead ofHTTPS.Locationheader is injected by Kong if the field is set to 301, 302, 307 or 308. Note: This config applies only if the route is configured to only accept thehttpsprotocol. Default: 426 - methods Sequence[str]
- A list of HTTP methods that match this route.
- model
Ai
Gateway Model Api Config Route Model - Configuration for overriding routing to this model using a selector. When no selector location is set, the format default selector is used. When values are not set, the model name is used as the selector value.
- paths Sequence[str]
- A list of paths that match this route.
- preserve_
host bool - When matching a route via one of the
hostsdomain names, use the requestHostheader in the upstream request headers. If set tofalse, the upstreamHostheader will be that of the service'shost. Default: false - protocols Sequence[str]
- An array of the protocols this route should allow. See the route Object section for a list of accepted protocols. When set to only
https, HTTP requests are answered with an upgrade error. When set to onlyhttp, HTTPS requests are answered with an error. Default: ["http","https"] - regex_
priority float - A number used to choose which route resolves a given request when several routes match it using regexes simultaneously. When two routes match the path and have the same
regex_priority, the older one (lowestcreated_at) is used. Note that the priority for non-regex routes is different (longer non-regex routes are matched before shorter ones). Default: 0 - request_
buffering bool - Whether to enable request body buffering or not. With HTTP 1.1, it may make sense to turn this off on services that receive data with chunked transfer encoding. Default: true
- response_
buffering bool - Whether to enable response body buffering or not. With HTTP 1.1, it may make sense to turn this off on services that send data with chunked transfer encoding. Default: true
- strip_
path bool - When matching a route via one of the
paths, strip the matching prefix from the upstream request URL. Default: true - Sequence[str]
- An optional set of strings associated with the route for grouping and filtering.
- headers Map<String>
- One or more lists of values indexed by header name that will cause this route to match if present in the request. The
Hostheader cannot be used with this attribute: hosts should be specified using thehostsattribute. Whenheaderscontains only one value and that value starts with the special prefix~*, the value is interpreted as a regular expression. - hosts List<String>
- A list of domain names that match this route. Note that the hosts value is case sensitive.
- https
Redirect NumberStatus Code - The status code Kong responds with when all properties of a route match except the protocol i.e. if the protocol of the request is
HTTPinstead ofHTTPS.Locationheader is injected by Kong if the field is set to 301, 302, 307 or 308. Note: This config applies only if the route is configured to only accept thehttpsprotocol. Default: 426 - methods List<String>
- A list of HTTP methods that match this route.
- model Property Map
- Configuration for overriding routing to this model using a selector. When no selector location is set, the format default selector is used. When values are not set, the model name is used as the selector value.
- paths List<String>
- A list of paths that match this route.
- preserve
Host Boolean - When matching a route via one of the
hostsdomain names, use the requestHostheader in the upstream request headers. If set tofalse, the upstreamHostheader will be that of the service'shost. Default: false - protocols List<String>
- An array of the protocols this route should allow. See the route Object section for a list of accepted protocols. When set to only
https, HTTP requests are answered with an upgrade error. When set to onlyhttp, HTTPS requests are answered with an error. Default: ["http","https"] - regex
Priority Number - A number used to choose which route resolves a given request when several routes match it using regexes simultaneously. When two routes match the path and have the same
regex_priority, the older one (lowestcreated_at) is used. Note that the priority for non-regex routes is different (longer non-regex routes are matched before shorter ones). Default: 0 - request
Buffering Boolean - Whether to enable request body buffering or not. With HTTP 1.1, it may make sense to turn this off on services that receive data with chunked transfer encoding. Default: true
- response
Buffering Boolean - Whether to enable response body buffering or not. With HTTP 1.1, it may make sense to turn this off on services that send data with chunked transfer encoding. Default: true
- strip
Path Boolean - When matching a route via one of the
paths, strip the matching prefix from the upstream request URL. Default: true - List<String>
- An optional set of strings associated with the route for grouping and filtering.
AiGatewayModelApiConfigRouteModel, AiGatewayModelApiConfigRouteModelArgs
- Body
Param string - The body property name to match for routing.
- Header
Param string - The header property name to match for routing.
- Path
Param string - The name of the regex capture group defined in the route path for routing.
- Values List<string>
- An optional model alias. When omitted, the model name is used. When no selector location is configured, the format default selector is used.
- Body
Param string - The body property name to match for routing.
- Header
Param string - The header property name to match for routing.
- Path
Param string - The name of the regex capture group defined in the route path for routing.
- Values []string
- An optional model alias. When omitted, the model name is used. When no selector location is configured, the format default selector is used.
- body_
param string - The body property name to match for routing.
- header_
param string - The header property name to match for routing.
- path_
param string - The name of the regex capture group defined in the route path for routing.
- values list(string)
- An optional model alias. When omitted, the model name is used. When no selector location is configured, the format default selector is used.
- body
Param String - The body property name to match for routing.
- header
Param String - The header property name to match for routing.
- path
Param String - The name of the regex capture group defined in the route path for routing.
- values List<String>
- An optional model alias. When omitted, the model name is used. When no selector location is configured, the format default selector is used.
- body
Param string - The body property name to match for routing.
- header
Param string - The header property name to match for routing.
- path
Param string - The name of the regex capture group defined in the route path for routing.
- values string[]
- An optional model alias. When omitted, the model name is used. When no selector location is configured, the format default selector is used.
- body_
param str - The body property name to match for routing.
- header_
param str - The header property name to match for routing.
- path_
param str - The name of the regex capture group defined in the route path for routing.
- values Sequence[str]
- An optional model alias. When omitted, the model name is used. When no selector location is configured, the format default selector is used.
- body
Param String - The body property name to match for routing.
- header
Param String - The header property name to match for routing.
- path
Param String - The name of the regex capture group defined in the route path for routing.
- values List<String>
- An optional model alias. When omitted, the model name is used. When no selector location is configured, the format default selector is used.
AiGatewayModelApiFormat, AiGatewayModelApiFormatArgs
- Type string
- The format type. possible known values include one of ["anthropic", "bedrock", "cohere", "gemini", "huggingface", "openai"]
- Type string
- The format type. possible known values include one of ["anthropic", "bedrock", "cohere", "gemini", "huggingface", "openai"]
- type string
- The format type. possible known values include one of ["anthropic", "bedrock", "cohere", "gemini", "huggingface", "openai"]
- type String
- The format type. possible known values include one of ["anthropic", "bedrock", "cohere", "gemini", "huggingface", "openai"]
- type string
- The format type. possible known values include one of ["anthropic", "bedrock", "cohere", "gemini", "huggingface", "openai"]
- type str
- The format type. possible known values include one of ["anthropic", "bedrock", "cohere", "gemini", "huggingface", "openai"]
- type String
- The format type. possible known values include one of ["anthropic", "bedrock", "cohere", "gemini", "huggingface", "openai"]
AiGatewayModelApiTarget, AiGatewayModelApiTargetArgs
- Allow
Auth boolOverride - When enabled, request-level auth parameters (such as API keys or bearer tokens) will override the static values defined for the provider. Default: false
- Config
Ai
Gateway Model Api Target Config - Configuration for a target model. Not Null
- Name string
- The name of the model defined in the upstream provider that will be executed. Not Null
- Provider string
- Reference to a model provider instance by name. Not Null
- Semantic
Description string - The semantic description of the target, required if using semantic load balancing. Specially, setting this to 'CATCHALL' will indicate such target to be used when no other targets match the semantic threshold.
- Weight double
- The weight this target gets within the upstream load balancer. Default: 100
- Allow
Auth boolOverride - When enabled, request-level auth parameters (such as API keys or bearer tokens) will override the static values defined for the provider. Default: false
- Config
Ai
Gateway Model Api Target Config - Configuration for a target model. Not Null
- Name string
- The name of the model defined in the upstream provider that will be executed. Not Null
- Provider string
- Reference to a model provider instance by name. Not Null
- Semantic
Description string - The semantic description of the target, required if using semantic load balancing. Specially, setting this to 'CATCHALL' will indicate such target to be used when no other targets match the semantic threshold.
- Weight float64
- The weight this target gets within the upstream load balancer. Default: 100
- allow_
auth_ booloverride - When enabled, request-level auth parameters (such as API keys or bearer tokens) will override the static values defined for the provider. Default: false
- config object
- Configuration for a target model. Not Null
- name string
- The name of the model defined in the upstream provider that will be executed. Not Null
- provider string
- Reference to a model provider instance by name. Not Null
- semantic_
description string - The semantic description of the target, required if using semantic load balancing. Specially, setting this to 'CATCHALL' will indicate such target to be used when no other targets match the semantic threshold.
- weight number
- The weight this target gets within the upstream load balancer. Default: 100
- allow
Auth BooleanOverride - When enabled, request-level auth parameters (such as API keys or bearer tokens) will override the static values defined for the provider. Default: false
- config
Ai
Gateway Model Api Target Config - Configuration for a target model. Not Null
- name String
- The name of the model defined in the upstream provider that will be executed. Not Null
- provider String
- Reference to a model provider instance by name. Not Null
- semantic
Description String - The semantic description of the target, required if using semantic load balancing. Specially, setting this to 'CATCHALL' will indicate such target to be used when no other targets match the semantic threshold.
- weight Double
- The weight this target gets within the upstream load balancer. Default: 100
- allow
Auth booleanOverride - When enabled, request-level auth parameters (such as API keys or bearer tokens) will override the static values defined for the provider. Default: false
- config
Ai
Gateway Model Api Target Config - Configuration for a target model. Not Null
- name string
- The name of the model defined in the upstream provider that will be executed. Not Null
- provider string
- Reference to a model provider instance by name. Not Null
- semantic
Description string - The semantic description of the target, required if using semantic load balancing. Specially, setting this to 'CATCHALL' will indicate such target to be used when no other targets match the semantic threshold.
- weight number
- The weight this target gets within the upstream load balancer. Default: 100
- allow_
auth_ booloverride - When enabled, request-level auth parameters (such as API keys or bearer tokens) will override the static values defined for the provider. Default: false
- config
Ai
Gateway Model Api Target Config - Configuration for a target model. Not Null
- name str
- The name of the model defined in the upstream provider that will be executed. Not Null
- provider str
- Reference to a model provider instance by name. Not Null
- semantic_
description str - The semantic description of the target, required if using semantic load balancing. Specially, setting this to 'CATCHALL' will indicate such target to be used when no other targets match the semantic threshold.
- weight float
- The weight this target gets within the upstream load balancer. Default: 100
- allow
Auth BooleanOverride - When enabled, request-level auth parameters (such as API keys or bearer tokens) will override the static values defined for the provider. Default: false
- config Property Map
- Configuration for a target model. Not Null
- name String
- The name of the model defined in the upstream provider that will be executed. Not Null
- provider String
- Reference to a model provider instance by name. Not Null
- semantic
Description String - The semantic description of the target, required if using semantic load balancing. Specially, setting this to 'CATCHALL' will indicate such target to be used when no other targets match the semantic threshold.
- weight Number
- The weight this target gets within the upstream load balancer. Default: 100
AiGatewayModelApiTargetConfig, AiGatewayModelApiTargetConfigArgs
- Anthropic
Ai
Gateway Model Api Target Config Anthropic - Anthropic-specific configuration for a model.
- Azure
Ai
Gateway Model Api Target Config Azure - Azure-specific configuration for a model.
- Bedrock
Ai
Gateway Model Api Target Config Bedrock - AWS Bedrock-specific configuration for a model.
- Cerebras
Ai
Gateway Model Api Target Config Cerebras - Cerebras-specific configuration for a model.
- Cohere
Ai
Gateway Model Api Target Config Cohere - Cohere-specific configuration for a model.
- Dashscope
Ai
Gateway Model Api Target Config Dashscope - Alibaba DashScope-specific configuration for a model.
- Databricks
Ai
Gateway Model Api Target Config Databricks - Databricks-specific configuration for a model.
- Deepseek
Ai
Gateway Model Api Target Config Deepseek - Deepseek-specific configuration for a model.
- Gemini
Ai
Gateway Model Api Target Config Gemini - Google Gemini-specific configuration for a model.
- Huggingface
Ai
Gateway Model Api Target Config Huggingface - Hugging Face-specific configuration for a model.
- Kimi
Ai
Gateway Model Api Target Config Kimi - Kimi (Moonshot AI)-specific configuration for a model.
- Llama2
Ai
Gateway Model Api Target Config Llama2 - Llama2-specific configuration for a model.
- Mistral
Ai
Gateway Model Api Target Config Mistral - Mistral-specific configuration for a model.
- Ollama
Ai
Gateway Model Api Target Config Ollama - Ollama-specific configuration for a model.
- Openai
Ai
Gateway Model Api Target Config Openai - Openai-specific configuration for a model.
- Sagemaker
Ai
Gateway Model Api Target Config Sagemaker - AWS SageMaker-specific configuration for a model.
- Vercel
Ai
Gateway Model Api Target Config Vercel - Vercel AI Gateway-specific configuration for a model.
- Vllm
Ai
Gateway Model Api Target Config Vllm - Vllm-specific configuration for a model.
- Xai
Ai
Gateway Model Api Target Config Xai - Xai-specific configuration for a model.
- Anthropic
Ai
Gateway Model Api Target Config Anthropic - Anthropic-specific configuration for a model.
- Azure
Ai
Gateway Model Api Target Config Azure - Azure-specific configuration for a model.
- Bedrock
Ai
Gateway Model Api Target Config Bedrock - AWS Bedrock-specific configuration for a model.
- Cerebras
Ai
Gateway Model Api Target Config Cerebras - Cerebras-specific configuration for a model.
- Cohere
Ai
Gateway Model Api Target Config Cohere - Cohere-specific configuration for a model.
- Dashscope
Ai
Gateway Model Api Target Config Dashscope - Alibaba DashScope-specific configuration for a model.
- Databricks
Ai
Gateway Model Api Target Config Databricks - Databricks-specific configuration for a model.
- Deepseek
Ai
Gateway Model Api Target Config Deepseek - Deepseek-specific configuration for a model.
- Gemini
Ai
Gateway Model Api Target Config Gemini - Google Gemini-specific configuration for a model.
- Huggingface
Ai
Gateway Model Api Target Config Huggingface - Hugging Face-specific configuration for a model.
- Kimi
Ai
Gateway Model Api Target Config Kimi - Kimi (Moonshot AI)-specific configuration for a model.
- Llama2
Ai
Gateway Model Api Target Config Llama2 - Llama2-specific configuration for a model.
- Mistral
Ai
Gateway Model Api Target Config Mistral - Mistral-specific configuration for a model.
- Ollama
Ai
Gateway Model Api Target Config Ollama - Ollama-specific configuration for a model.
- Openai
Ai
Gateway Model Api Target Config Openai - Openai-specific configuration for a model.
- Sagemaker
Ai
Gateway Model Api Target Config Sagemaker - AWS SageMaker-specific configuration for a model.
- Vercel
Ai
Gateway Model Api Target Config Vercel - Vercel AI Gateway-specific configuration for a model.
- Vllm
Ai
Gateway Model Api Target Config Vllm - Vllm-specific configuration for a model.
- Xai
Ai
Gateway Model Api Target Config Xai - Xai-specific configuration for a model.
- anthropic object
- Anthropic-specific configuration for a model.
- azure object
- Azure-specific configuration for a model.
- bedrock object
- AWS Bedrock-specific configuration for a model.
- cerebras object
- Cerebras-specific configuration for a model.
- cohere object
- Cohere-specific configuration for a model.
- dashscope object
- Alibaba DashScope-specific configuration for a model.
- databricks object
- Databricks-specific configuration for a model.
- deepseek object
- Deepseek-specific configuration for a model.
- gemini object
- Google Gemini-specific configuration for a model.
- huggingface object
- Hugging Face-specific configuration for a model.
- kimi object
- Kimi (Moonshot AI)-specific configuration for a model.
- llama2 object
- Llama2-specific configuration for a model.
- mistral object
- Mistral-specific configuration for a model.
- ollama object
- Ollama-specific configuration for a model.
- openai object
- Openai-specific configuration for a model.
- sagemaker object
- AWS SageMaker-specific configuration for a model.
- vercel object
- Vercel AI Gateway-specific configuration for a model.
- vllm object
- Vllm-specific configuration for a model.
- xai object
- Xai-specific configuration for a model.
- anthropic
Ai
Gateway Model Api Target Config Anthropic - Anthropic-specific configuration for a model.
- azure
Ai
Gateway Model Api Target Config Azure - Azure-specific configuration for a model.
- bedrock
Ai
Gateway Model Api Target Config Bedrock - AWS Bedrock-specific configuration for a model.
- cerebras
Ai
Gateway Model Api Target Config Cerebras - Cerebras-specific configuration for a model.
- cohere
Ai
Gateway Model Api Target Config Cohere - Cohere-specific configuration for a model.
- dashscope
Ai
Gateway Model Api Target Config Dashscope - Alibaba DashScope-specific configuration for a model.
- databricks
Ai
Gateway Model Api Target Config Databricks - Databricks-specific configuration for a model.
- deepseek
Ai
Gateway Model Api Target Config Deepseek - Deepseek-specific configuration for a model.
- gemini
Ai
Gateway Model Api Target Config Gemini - Google Gemini-specific configuration for a model.
- huggingface
Ai
Gateway Model Api Target Config Huggingface - Hugging Face-specific configuration for a model.
- kimi
Ai
Gateway Model Api Target Config Kimi - Kimi (Moonshot AI)-specific configuration for a model.
- llama2
Ai
Gateway Model Api Target Config Llama2 - Llama2-specific configuration for a model.
- mistral
Ai
Gateway Model Api Target Config Mistral - Mistral-specific configuration for a model.
- ollama
Ai
Gateway Model Api Target Config Ollama - Ollama-specific configuration for a model.
- openai
Ai
Gateway Model Api Target Config Openai - Openai-specific configuration for a model.
- sagemaker
Ai
Gateway Model Api Target Config Sagemaker - AWS SageMaker-specific configuration for a model.
- vercel
Ai
Gateway Model Api Target Config Vercel - Vercel AI Gateway-specific configuration for a model.
- vllm
Ai
Gateway Model Api Target Config Vllm - Vllm-specific configuration for a model.
- xai
Ai
Gateway Model Api Target Config Xai - Xai-specific configuration for a model.
- anthropic
Ai
Gateway Model Api Target Config Anthropic - Anthropic-specific configuration for a model.
- azure
Ai
Gateway Model Api Target Config Azure - Azure-specific configuration for a model.
- bedrock
Ai
Gateway Model Api Target Config Bedrock - AWS Bedrock-specific configuration for a model.
- cerebras
Ai
Gateway Model Api Target Config Cerebras - Cerebras-specific configuration for a model.
- cohere
Ai
Gateway Model Api Target Config Cohere - Cohere-specific configuration for a model.
- dashscope
Ai
Gateway Model Api Target Config Dashscope - Alibaba DashScope-specific configuration for a model.
- databricks
Ai
Gateway Model Api Target Config Databricks - Databricks-specific configuration for a model.
- deepseek
Ai
Gateway Model Api Target Config Deepseek - Deepseek-specific configuration for a model.
- gemini
Ai
Gateway Model Api Target Config Gemini - Google Gemini-specific configuration for a model.
- huggingface
Ai
Gateway Model Api Target Config Huggingface - Hugging Face-specific configuration for a model.
- kimi
Ai
Gateway Model Api Target Config Kimi - Kimi (Moonshot AI)-specific configuration for a model.
- llama2
Ai
Gateway Model Api Target Config Llama2 - Llama2-specific configuration for a model.
- mistral
Ai
Gateway Model Api Target Config Mistral - Mistral-specific configuration for a model.
- ollama
Ai
Gateway Model Api Target Config Ollama - Ollama-specific configuration for a model.
- openai
Ai
Gateway Model Api Target Config Openai - Openai-specific configuration for a model.
- sagemaker
Ai
Gateway Model Api Target Config Sagemaker - AWS SageMaker-specific configuration for a model.
- vercel
Ai
Gateway Model Api Target Config Vercel - Vercel AI Gateway-specific configuration for a model.
- vllm
Ai
Gateway Model Api Target Config Vllm - Vllm-specific configuration for a model.
- xai
Ai
Gateway Model Api Target Config Xai - Xai-specific configuration for a model.
- anthropic
Ai
Gateway Model Api Target Config Anthropic - Anthropic-specific configuration for a model.
- azure
Ai
Gateway Model Api Target Config Azure - Azure-specific configuration for a model.
- bedrock
Ai
Gateway Model Api Target Config Bedrock - AWS Bedrock-specific configuration for a model.
- cerebras
Ai
Gateway Model Api Target Config Cerebras - Cerebras-specific configuration for a model.
- cohere
Ai
Gateway Model Api Target Config Cohere - Cohere-specific configuration for a model.
- dashscope
Ai
Gateway Model Api Target Config Dashscope - Alibaba DashScope-specific configuration for a model.
- databricks
Ai
Gateway Model Api Target Config Databricks - Databricks-specific configuration for a model.
- deepseek
Ai
Gateway Model Api Target Config Deepseek - Deepseek-specific configuration for a model.
- gemini
Ai
Gateway Model Api Target Config Gemini - Google Gemini-specific configuration for a model.
- huggingface
Ai
Gateway Model Api Target Config Huggingface - Hugging Face-specific configuration for a model.
- kimi
Ai
Gateway Model Api Target Config Kimi - Kimi (Moonshot AI)-specific configuration for a model.
- llama2
Ai
Gateway Model Api Target Config Llama2 - Llama2-specific configuration for a model.
- mistral
Ai
Gateway Model Api Target Config Mistral - Mistral-specific configuration for a model.
- ollama
Ai
Gateway Model Api Target Config Ollama - Ollama-specific configuration for a model.
- openai
Ai
Gateway Model Api Target Config Openai - Openai-specific configuration for a model.
- sagemaker
Ai
Gateway Model Api Target Config Sagemaker - AWS SageMaker-specific configuration for a model.
- vercel
Ai
Gateway Model Api Target Config Vercel - Vercel AI Gateway-specific configuration for a model.
- vllm
Ai
Gateway Model Api Target Config Vllm - Vllm-specific configuration for a model.
- xai
Ai
Gateway Model Api Target Config Xai - Xai-specific configuration for a model.
- anthropic Property Map
- Anthropic-specific configuration for a model.
- azure Property Map
- Azure-specific configuration for a model.
- bedrock Property Map
- AWS Bedrock-specific configuration for a model.
- cerebras Property Map
- Cerebras-specific configuration for a model.
- cohere Property Map
- Cohere-specific configuration for a model.
- dashscope Property Map
- Alibaba DashScope-specific configuration for a model.
- databricks Property Map
- Databricks-specific configuration for a model.
- deepseek Property Map
- Deepseek-specific configuration for a model.
- gemini Property Map
- Google Gemini-specific configuration for a model.
- huggingface Property Map
- Hugging Face-specific configuration for a model.
- kimi Property Map
- Kimi (Moonshot AI)-specific configuration for a model.
- llama2 Property Map
- Llama2-specific configuration for a model.
- mistral Property Map
- Mistral-specific configuration for a model.
- ollama Property Map
- Ollama-specific configuration for a model.
- openai Property Map
- Openai-specific configuration for a model.
- sagemaker Property Map
- AWS SageMaker-specific configuration for a model.
- vercel Property Map
- Vercel AI Gateway-specific configuration for a model.
- vllm Property Map
- Vllm-specific configuration for a model.
- xai Property Map
- Xai-specific configuration for a model.
AiGatewayModelApiTargetConfigAnthropic, AiGatewayModelApiTargetConfigAnthropicArgs
- Cache
Read doubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write doubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write List<AiCost Lists Gateway Model Api Target Config Anthropic Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window List<AiFactors Gateway Model Api Target Config Anthropic Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions double - The number of dimensions for embedding outputs.
- Input
Cost double - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens double - The maximum number of tokens to generate in the response.
- Output
Cost double - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier List<AiFactors Gateway Model Api Target Config Anthropic Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature double
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K double - Limits the number of highest-probability tokens considered during generation.
- Top
P double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- Version string
- The Anthropic API version to use. Default: "2023-06-01"
- Cache
Read float64Cost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write float64Cost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write []AiCost Lists Gateway Model Api Target Config Anthropic Cache Write Cost List - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window []AiFactors Gateway Model Api Target Config Anthropic Context Window Factor - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions float64 - The number of dimensions for embedding outputs.
- Input
Cost float64 - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens float64 - The maximum number of tokens to generate in the response.
- Output
Cost float64 - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier []AiFactors Gateway Model Api Target Config Anthropic Service Tier Factor - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature float64
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K float64 - Limits the number of highest-probability tokens considered during generation.
- Top
P float64 - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- Version string
- The Anthropic API version to use. Default: "2023-06-01"
- cache_
read_ numbercost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ numbercost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ list(object)cost_ lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ list(object)factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions number - The number of dimensions for embedding outputs.
- input_
cost number - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens number - The maximum number of tokens to generate in the response.
- output_
cost number - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ list(object)factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k number - Limits the number of highest-probability tokens considered during generation.
- top_
p number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url string - The upstream URL for the model endpoint.
- version string
- The Anthropic API version to use. Default: "2023-06-01"
- cache
Read DoubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write DoubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<AiCost Lists Gateway Model Api Target Config Anthropic Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<AiFactors Gateway Model Api Target Config Anthropic Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Double - The number of dimensions for embedding outputs.
- input
Cost Double - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Double - The maximum number of tokens to generate in the response.
- output
Cost Double - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<AiFactors Gateway Model Api Target Config Anthropic Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Double
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Double - Limits the number of highest-probability tokens considered during generation.
- top
P Double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
- version String
- The Anthropic API version to use. Default: "2023-06-01"
- cache
Read numberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write numberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write AiCost Lists Gateway Model Api Target Config Anthropic Cache Write Cost List[] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window AiFactors Gateway Model Api Target Config Anthropic Context Window Factor[] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions number - The number of dimensions for embedding outputs.
- input
Cost number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens number - The maximum number of tokens to generate in the response.
- output
Cost number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier AiFactors Gateway Model Api Target Config Anthropic Service Tier Factor[] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K number - Limits the number of highest-probability tokens considered during generation.
- top
P number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url string - The upstream URL for the model endpoint.
- version string
- The Anthropic API version to use. Default: "2023-06-01"
- cache_
read_ floatcost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ floatcost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ Sequence[Aicost_ lists Gateway Model Api Target Config Anthropic Cache Write Cost List] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ Sequence[Aifactors Gateway Model Api Target Config Anthropic Context Window Factor] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions float - The number of dimensions for embedding outputs.
- input_
cost float - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens float - The maximum number of tokens to generate in the response.
- output_
cost float - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ Sequence[Aifactors Gateway Model Api Target Config Anthropic Service Tier Factor] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature float
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k float - Limits the number of highest-probability tokens considered during generation.
- top_
p float - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url str - The upstream URL for the model endpoint.
- version str
- The Anthropic API version to use. Default: "2023-06-01"
- cache
Read NumberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write NumberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<Property Map>Cost Lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<Property Map>Factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Number - The number of dimensions for embedding outputs.
- input
Cost Number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Number - The maximum number of tokens to generate in the response.
- output
Cost Number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<Property Map>Factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Number - Limits the number of highest-probability tokens considered during generation.
- top
P Number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
- version String
- The Anthropic API version to use. Default: "2023-06-01"
AiGatewayModelApiTargetConfigAnthropicCacheWriteCostList, AiGatewayModelApiTargetConfigAnthropicCacheWriteCostListArgs
AiGatewayModelApiTargetConfigAnthropicContextWindowFactor, AiGatewayModelApiTargetConfigAnthropicContextWindowFactorArgs
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor double - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor double - Multiplier applied to output pricing above the threshold. Not Null
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor float64 - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor float64 - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor number - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor number - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Double - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Double - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor number - Multiplier applied to output pricing above the threshold. Not Null
- above str
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor float - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor float - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Number - Multiplier applied to output pricing above the threshold. Not Null
AiGatewayModelApiTargetConfigAnthropicServiceTierFactor, AiGatewayModelApiTargetConfigAnthropicServiceTierFactorArgs
- Factor double
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- Factor float64
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Double
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor float
- Multiplier applied to the whole request for this service tier. Not Null
- tier str
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Number
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
AiGatewayModelApiTargetConfigAzure, AiGatewayModelApiTargetConfigAzureArgs
- Api
Version string - The Azure OpenAI API version to use. Default: "2023-05-15"
- Cache
Read doubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write doubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write List<AiCost Lists Gateway Model Api Target Config Azure Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window List<AiFactors Gateway Model Api Target Config Azure Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Deployment
Id string - The Azure deployment ID for the model. Applies when the Azure provider's
serviceisazure-openai; not used forazure-foundry. - Embeddings
Dimensions double - The number of dimensions for embedding outputs.
- Foundry
Path stringPrefix - The API path prefix for the Azure AI Foundry endpoint, selecting the model's
API surface.
/openai/v1targets the OpenAI-compatible surface;/anthropic/v1targets the Anthropic surface. Applies when the Azure provider'sserviceisazure-foundry. possible known values include one of ["/openai/v1", "/anthropic/v1"]; Default: "/openai/v1" - Input
Cost double - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens double - The maximum number of tokens to generate in the response.
- Output
Cost double - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier List<AiFactors Gateway Model Api Target Config Azure Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature double
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K double - Limits the number of highest-probability tokens considered during generation.
- Top
P double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- Api
Version string - The Azure OpenAI API version to use. Default: "2023-05-15"
- Cache
Read float64Cost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write float64Cost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write []AiCost Lists Gateway Model Api Target Config Azure Cache Write Cost List - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window []AiFactors Gateway Model Api Target Config Azure Context Window Factor - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Deployment
Id string - The Azure deployment ID for the model. Applies when the Azure provider's
serviceisazure-openai; not used forazure-foundry. - Embeddings
Dimensions float64 - The number of dimensions for embedding outputs.
- Foundry
Path stringPrefix - The API path prefix for the Azure AI Foundry endpoint, selecting the model's
API surface.
/openai/v1targets the OpenAI-compatible surface;/anthropic/v1targets the Anthropic surface. Applies when the Azure provider'sserviceisazure-foundry. possible known values include one of ["/openai/v1", "/anthropic/v1"]; Default: "/openai/v1" - Input
Cost float64 - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens float64 - The maximum number of tokens to generate in the response.
- Output
Cost float64 - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier []AiFactors Gateway Model Api Target Config Azure Service Tier Factor - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature float64
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K float64 - Limits the number of highest-probability tokens considered during generation.
- Top
P float64 - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- api_
version string - The Azure OpenAI API version to use. Default: "2023-05-15"
- cache_
read_ numbercost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ numbercost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ list(object)cost_ lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ list(object)factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- deployment_
id string - The Azure deployment ID for the model. Applies when the Azure provider's
serviceisazure-openai; not used forazure-foundry. - embeddings_
dimensions number - The number of dimensions for embedding outputs.
- foundry_
path_ stringprefix - The API path prefix for the Azure AI Foundry endpoint, selecting the model's
API surface.
/openai/v1targets the OpenAI-compatible surface;/anthropic/v1targets the Anthropic surface. Applies when the Azure provider'sserviceisazure-foundry. possible known values include one of ["/openai/v1", "/anthropic/v1"]; Default: "/openai/v1" - input_
cost number - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens number - The maximum number of tokens to generate in the response.
- output_
cost number - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ list(object)factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k number - Limits the number of highest-probability tokens considered during generation.
- top_
p number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url string - The upstream URL for the model endpoint.
- api
Version String - The Azure OpenAI API version to use. Default: "2023-05-15"
- cache
Read DoubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write DoubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<AiCost Lists Gateway Model Api Target Config Azure Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<AiFactors Gateway Model Api Target Config Azure Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- deployment
Id String - The Azure deployment ID for the model. Applies when the Azure provider's
serviceisazure-openai; not used forazure-foundry. - embeddings
Dimensions Double - The number of dimensions for embedding outputs.
- foundry
Path StringPrefix - The API path prefix for the Azure AI Foundry endpoint, selecting the model's
API surface.
/openai/v1targets the OpenAI-compatible surface;/anthropic/v1targets the Anthropic surface. Applies when the Azure provider'sserviceisazure-foundry. possible known values include one of ["/openai/v1", "/anthropic/v1"]; Default: "/openai/v1" - input
Cost Double - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Double - The maximum number of tokens to generate in the response.
- output
Cost Double - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<AiFactors Gateway Model Api Target Config Azure Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Double
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Double - Limits the number of highest-probability tokens considered during generation.
- top
P Double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
- api
Version string - The Azure OpenAI API version to use. Default: "2023-05-15"
- cache
Read numberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write numberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write AiCost Lists Gateway Model Api Target Config Azure Cache Write Cost List[] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window AiFactors Gateway Model Api Target Config Azure Context Window Factor[] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- deployment
Id string - The Azure deployment ID for the model. Applies when the Azure provider's
serviceisazure-openai; not used forazure-foundry. - embeddings
Dimensions number - The number of dimensions for embedding outputs.
- foundry
Path stringPrefix - The API path prefix for the Azure AI Foundry endpoint, selecting the model's
API surface.
/openai/v1targets the OpenAI-compatible surface;/anthropic/v1targets the Anthropic surface. Applies when the Azure provider'sserviceisazure-foundry. possible known values include one of ["/openai/v1", "/anthropic/v1"]; Default: "/openai/v1" - input
Cost number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens number - The maximum number of tokens to generate in the response.
- output
Cost number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier AiFactors Gateway Model Api Target Config Azure Service Tier Factor[] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K number - Limits the number of highest-probability tokens considered during generation.
- top
P number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url string - The upstream URL for the model endpoint.
- api_
version str - The Azure OpenAI API version to use. Default: "2023-05-15"
- cache_
read_ floatcost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ floatcost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ Sequence[Aicost_ lists Gateway Model Api Target Config Azure Cache Write Cost List] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ Sequence[Aifactors Gateway Model Api Target Config Azure Context Window Factor] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- deployment_
id str - The Azure deployment ID for the model. Applies when the Azure provider's
serviceisazure-openai; not used forazure-foundry. - embeddings_
dimensions float - The number of dimensions for embedding outputs.
- foundry_
path_ strprefix - The API path prefix for the Azure AI Foundry endpoint, selecting the model's
API surface.
/openai/v1targets the OpenAI-compatible surface;/anthropic/v1targets the Anthropic surface. Applies when the Azure provider'sserviceisazure-foundry. possible known values include one of ["/openai/v1", "/anthropic/v1"]; Default: "/openai/v1" - input_
cost float - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens float - The maximum number of tokens to generate in the response.
- output_
cost float - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ Sequence[Aifactors Gateway Model Api Target Config Azure Service Tier Factor] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature float
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k float - Limits the number of highest-probability tokens considered during generation.
- top_
p float - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url str - The upstream URL for the model endpoint.
- api
Version String - The Azure OpenAI API version to use. Default: "2023-05-15"
- cache
Read NumberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write NumberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<Property Map>Cost Lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<Property Map>Factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- deployment
Id String - The Azure deployment ID for the model. Applies when the Azure provider's
serviceisazure-openai; not used forazure-foundry. - embeddings
Dimensions Number - The number of dimensions for embedding outputs.
- foundry
Path StringPrefix - The API path prefix for the Azure AI Foundry endpoint, selecting the model's
API surface.
/openai/v1targets the OpenAI-compatible surface;/anthropic/v1targets the Anthropic surface. Applies when the Azure provider'sserviceisazure-foundry. possible known values include one of ["/openai/v1", "/anthropic/v1"]; Default: "/openai/v1" - input
Cost Number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Number - The maximum number of tokens to generate in the response.
- output
Cost Number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<Property Map>Factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Number - Limits the number of highest-probability tokens considered during generation.
- top
P Number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
AiGatewayModelApiTargetConfigAzureCacheWriteCostList, AiGatewayModelApiTargetConfigAzureCacheWriteCostListArgs
AiGatewayModelApiTargetConfigAzureContextWindowFactor, AiGatewayModelApiTargetConfigAzureContextWindowFactorArgs
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor double - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor double - Multiplier applied to output pricing above the threshold. Not Null
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor float64 - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor float64 - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor number - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor number - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Double - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Double - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor number - Multiplier applied to output pricing above the threshold. Not Null
- above str
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor float - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor float - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Number - Multiplier applied to output pricing above the threshold. Not Null
AiGatewayModelApiTargetConfigAzureServiceTierFactor, AiGatewayModelApiTargetConfigAzureServiceTierFactorArgs
- Factor double
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- Factor float64
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Double
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor float
- Multiplier applied to the whole request for this service tier. Not Null
- tier str
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Number
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
AiGatewayModelApiTargetConfigBedrock, AiGatewayModelApiTargetConfigBedrockArgs
- Batch
Bucket stringPrefix - S3 bucket prefix for batch inference jobs.
- Cache
Read doubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write doubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write List<AiCost Lists Gateway Model Api Target Config Bedrock Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window List<AiFactors Gateway Model Api Target Config Bedrock Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions double - The number of dimensions for embedding outputs.
- Embeddings
Normalize bool - Whether to normalize embedding vectors in the response. Default: false
- Input
Cost double - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens double - The maximum number of tokens to generate in the response.
- Output
Cost double - Cost per 1M output tokens for billing and cost tracking.
- Performance
Config stringLatency - Latency performance configuration for the model invocation.
- Region string
- The AWS region for the model. Setting this option overrides the AWS_REGION environment variable.
- Service
Tier List<AiFactors Gateway Model Api Target Config Bedrock Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature double
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K double - Limits the number of highest-probability tokens considered during generation.
- Top
P double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- Video
Output stringS3Uri - S3 URI for storing video generation outputs.
- Batch
Bucket stringPrefix - S3 bucket prefix for batch inference jobs.
- Cache
Read float64Cost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write float64Cost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write []AiCost Lists Gateway Model Api Target Config Bedrock Cache Write Cost List - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window []AiFactors Gateway Model Api Target Config Bedrock Context Window Factor - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions float64 - The number of dimensions for embedding outputs.
- Embeddings
Normalize bool - Whether to normalize embedding vectors in the response. Default: false
- Input
Cost float64 - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens float64 - The maximum number of tokens to generate in the response.
- Output
Cost float64 - Cost per 1M output tokens for billing and cost tracking.
- Performance
Config stringLatency - Latency performance configuration for the model invocation.
- Region string
- The AWS region for the model. Setting this option overrides the AWS_REGION environment variable.
- Service
Tier []AiFactors Gateway Model Api Target Config Bedrock Service Tier Factor - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature float64
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K float64 - Limits the number of highest-probability tokens considered during generation.
- Top
P float64 - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- Video
Output stringS3Uri - S3 URI for storing video generation outputs.
- batch_
bucket_ stringprefix - S3 bucket prefix for batch inference jobs.
- cache_
read_ numbercost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ numbercost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ list(object)cost_ lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ list(object)factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions number - The number of dimensions for embedding outputs.
- embeddings_
normalize bool - Whether to normalize embedding vectors in the response. Default: false
- input_
cost number - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens number - The maximum number of tokens to generate in the response.
- output_
cost number - Cost per 1M output tokens for billing and cost tracking.
- performance_
config_ stringlatency - Latency performance configuration for the model invocation.
- region string
- The AWS region for the model. Setting this option overrides the AWS_REGION environment variable.
- service_
tier_ list(object)factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k number - Limits the number of highest-probability tokens considered during generation.
- top_
p number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url string - The upstream URL for the model endpoint.
- video_
output_ strings3_ uri - S3 URI for storing video generation outputs.
- batch
Bucket StringPrefix - S3 bucket prefix for batch inference jobs.
- cache
Read DoubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write DoubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<AiCost Lists Gateway Model Api Target Config Bedrock Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<AiFactors Gateway Model Api Target Config Bedrock Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Double - The number of dimensions for embedding outputs.
- embeddings
Normalize Boolean - Whether to normalize embedding vectors in the response. Default: false
- input
Cost Double - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Double - The maximum number of tokens to generate in the response.
- output
Cost Double - Cost per 1M output tokens for billing and cost tracking.
- performance
Config StringLatency - Latency performance configuration for the model invocation.
- region String
- The AWS region for the model. Setting this option overrides the AWS_REGION environment variable.
- service
Tier List<AiFactors Gateway Model Api Target Config Bedrock Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Double
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Double - Limits the number of highest-probability tokens considered during generation.
- top
P Double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
- video
Output StringS3Uri - S3 URI for storing video generation outputs.
- batch
Bucket stringPrefix - S3 bucket prefix for batch inference jobs.
- cache
Read numberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write numberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write AiCost Lists Gateway Model Api Target Config Bedrock Cache Write Cost List[] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window AiFactors Gateway Model Api Target Config Bedrock Context Window Factor[] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions number - The number of dimensions for embedding outputs.
- embeddings
Normalize boolean - Whether to normalize embedding vectors in the response. Default: false
- input
Cost number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens number - The maximum number of tokens to generate in the response.
- output
Cost number - Cost per 1M output tokens for billing and cost tracking.
- performance
Config stringLatency - Latency performance configuration for the model invocation.
- region string
- The AWS region for the model. Setting this option overrides the AWS_REGION environment variable.
- service
Tier AiFactors Gateway Model Api Target Config Bedrock Service Tier Factor[] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K number - Limits the number of highest-probability tokens considered during generation.
- top
P number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url string - The upstream URL for the model endpoint.
- video
Output stringS3Uri - S3 URI for storing video generation outputs.
- batch_
bucket_ strprefix - S3 bucket prefix for batch inference jobs.
- cache_
read_ floatcost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ floatcost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ Sequence[Aicost_ lists Gateway Model Api Target Config Bedrock Cache Write Cost List] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ Sequence[Aifactors Gateway Model Api Target Config Bedrock Context Window Factor] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions float - The number of dimensions for embedding outputs.
- embeddings_
normalize bool - Whether to normalize embedding vectors in the response. Default: false
- input_
cost float - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens float - The maximum number of tokens to generate in the response.
- output_
cost float - Cost per 1M output tokens for billing and cost tracking.
- performance_
config_ strlatency - Latency performance configuration for the model invocation.
- region str
- The AWS region for the model. Setting this option overrides the AWS_REGION environment variable.
- service_
tier_ Sequence[Aifactors Gateway Model Api Target Config Bedrock Service Tier Factor] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature float
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k float - Limits the number of highest-probability tokens considered during generation.
- top_
p float - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url str - The upstream URL for the model endpoint.
- video_
output_ strs3_ uri - S3 URI for storing video generation outputs.
- batch
Bucket StringPrefix - S3 bucket prefix for batch inference jobs.
- cache
Read NumberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write NumberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<Property Map>Cost Lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<Property Map>Factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Number - The number of dimensions for embedding outputs.
- embeddings
Normalize Boolean - Whether to normalize embedding vectors in the response. Default: false
- input
Cost Number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Number - The maximum number of tokens to generate in the response.
- output
Cost Number - Cost per 1M output tokens for billing and cost tracking.
- performance
Config StringLatency - Latency performance configuration for the model invocation.
- region String
- The AWS region for the model. Setting this option overrides the AWS_REGION environment variable.
- service
Tier List<Property Map>Factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Number - Limits the number of highest-probability tokens considered during generation.
- top
P Number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
- video
Output StringS3Uri - S3 URI for storing video generation outputs.
AiGatewayModelApiTargetConfigBedrockCacheWriteCostList, AiGatewayModelApiTargetConfigBedrockCacheWriteCostListArgs
AiGatewayModelApiTargetConfigBedrockContextWindowFactor, AiGatewayModelApiTargetConfigBedrockContextWindowFactorArgs
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor double - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor double - Multiplier applied to output pricing above the threshold. Not Null
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor float64 - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor float64 - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor number - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor number - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Double - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Double - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor number - Multiplier applied to output pricing above the threshold. Not Null
- above str
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor float - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor float - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Number - Multiplier applied to output pricing above the threshold. Not Null
AiGatewayModelApiTargetConfigBedrockServiceTierFactor, AiGatewayModelApiTargetConfigBedrockServiceTierFactorArgs
- Factor double
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- Factor float64
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Double
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor float
- Multiplier applied to the whole request for this service tier. Not Null
- tier str
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Number
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
AiGatewayModelApiTargetConfigCerebras, AiGatewayModelApiTargetConfigCerebrasArgs
- Cache
Read doubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write doubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write List<AiCost Lists Gateway Model Api Target Config Cerebras Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window List<AiFactors Gateway Model Api Target Config Cerebras Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions double - The number of dimensions for embedding outputs.
- Input
Cost double - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens double - The maximum number of tokens to generate in the response.
- Output
Cost double - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier List<AiFactors Gateway Model Api Target Config Cerebras Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature double
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K double - Limits the number of highest-probability tokens considered during generation.
- Top
P double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- Cache
Read float64Cost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write float64Cost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write []AiCost Lists Gateway Model Api Target Config Cerebras Cache Write Cost List - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window []AiFactors Gateway Model Api Target Config Cerebras Context Window Factor - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions float64 - The number of dimensions for embedding outputs.
- Input
Cost float64 - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens float64 - The maximum number of tokens to generate in the response.
- Output
Cost float64 - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier []AiFactors Gateway Model Api Target Config Cerebras Service Tier Factor - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature float64
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K float64 - Limits the number of highest-probability tokens considered during generation.
- Top
P float64 - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- cache_
read_ numbercost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ numbercost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ list(object)cost_ lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ list(object)factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions number - The number of dimensions for embedding outputs.
- input_
cost number - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens number - The maximum number of tokens to generate in the response.
- output_
cost number - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ list(object)factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k number - Limits the number of highest-probability tokens considered during generation.
- top_
p number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url string - The upstream URL for the model endpoint.
- cache
Read DoubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write DoubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<AiCost Lists Gateway Model Api Target Config Cerebras Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<AiFactors Gateway Model Api Target Config Cerebras Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Double - The number of dimensions for embedding outputs.
- input
Cost Double - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Double - The maximum number of tokens to generate in the response.
- output
Cost Double - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<AiFactors Gateway Model Api Target Config Cerebras Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Double
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Double - Limits the number of highest-probability tokens considered during generation.
- top
P Double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
- cache
Read numberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write numberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write AiCost Lists Gateway Model Api Target Config Cerebras Cache Write Cost List[] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window AiFactors Gateway Model Api Target Config Cerebras Context Window Factor[] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions number - The number of dimensions for embedding outputs.
- input
Cost number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens number - The maximum number of tokens to generate in the response.
- output
Cost number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier AiFactors Gateway Model Api Target Config Cerebras Service Tier Factor[] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K number - Limits the number of highest-probability tokens considered during generation.
- top
P number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url string - The upstream URL for the model endpoint.
- cache_
read_ floatcost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ floatcost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ Sequence[Aicost_ lists Gateway Model Api Target Config Cerebras Cache Write Cost List] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ Sequence[Aifactors Gateway Model Api Target Config Cerebras Context Window Factor] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions float - The number of dimensions for embedding outputs.
- input_
cost float - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens float - The maximum number of tokens to generate in the response.
- output_
cost float - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ Sequence[Aifactors Gateway Model Api Target Config Cerebras Service Tier Factor] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature float
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k float - Limits the number of highest-probability tokens considered during generation.
- top_
p float - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url str - The upstream URL for the model endpoint.
- cache
Read NumberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write NumberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<Property Map>Cost Lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<Property Map>Factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Number - The number of dimensions for embedding outputs.
- input
Cost Number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Number - The maximum number of tokens to generate in the response.
- output
Cost Number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<Property Map>Factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Number - Limits the number of highest-probability tokens considered during generation.
- top
P Number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
AiGatewayModelApiTargetConfigCerebrasCacheWriteCostList, AiGatewayModelApiTargetConfigCerebrasCacheWriteCostListArgs
AiGatewayModelApiTargetConfigCerebrasContextWindowFactor, AiGatewayModelApiTargetConfigCerebrasContextWindowFactorArgs
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor double - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor double - Multiplier applied to output pricing above the threshold. Not Null
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor float64 - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor float64 - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor number - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor number - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Double - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Double - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor number - Multiplier applied to output pricing above the threshold. Not Null
- above str
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor float - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor float - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Number - Multiplier applied to output pricing above the threshold. Not Null
AiGatewayModelApiTargetConfigCerebrasServiceTierFactor, AiGatewayModelApiTargetConfigCerebrasServiceTierFactorArgs
- Factor double
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- Factor float64
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Double
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor float
- Multiplier applied to the whole request for this service tier. Not Null
- tier str
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Number
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
AiGatewayModelApiTargetConfigCohere, AiGatewayModelApiTargetConfigCohereArgs
- Api
Version string - Cohere API version.
v1uses the legacy/v1/chatendpoint;v2(default) uses/v2/chatand supports tool calling. possible known values include one of ["v1", "v2"]; Default: "v2" - Cache
Read doubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write doubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write List<AiCost Lists Gateway Model Api Target Config Cohere Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window List<AiFactors Gateway Model Api Target Config Cohere Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embedding
Input stringType - The intended downstream use of the embeddings to improve model quality. possible known values include one of ["classification", "clustering", "image", "searchdocument", "searchquery"]; Default: "classification"
- Embeddings
Dimensions double - The number of dimensions for embedding outputs.
- Input
Cost double - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens double - The maximum number of tokens to generate in the response.
- Output
Cost double - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier List<AiFactors Gateway Model Api Target Config Cohere Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature double
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K double - Limits the number of highest-probability tokens considered during generation.
- Top
P double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- Wait
For boolModel - Whether to wait for the model to be ready before sending the request. Default: false
- Api
Version string - Cohere API version.
v1uses the legacy/v1/chatendpoint;v2(default) uses/v2/chatand supports tool calling. possible known values include one of ["v1", "v2"]; Default: "v2" - Cache
Read float64Cost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write float64Cost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write []AiCost Lists Gateway Model Api Target Config Cohere Cache Write Cost List - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window []AiFactors Gateway Model Api Target Config Cohere Context Window Factor - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embedding
Input stringType - The intended downstream use of the embeddings to improve model quality. possible known values include one of ["classification", "clustering", "image", "searchdocument", "searchquery"]; Default: "classification"
- Embeddings
Dimensions float64 - The number of dimensions for embedding outputs.
- Input
Cost float64 - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens float64 - The maximum number of tokens to generate in the response.
- Output
Cost float64 - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier []AiFactors Gateway Model Api Target Config Cohere Service Tier Factor - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature float64
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K float64 - Limits the number of highest-probability tokens considered during generation.
- Top
P float64 - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- Wait
For boolModel - Whether to wait for the model to be ready before sending the request. Default: false
- api_
version string - Cohere API version.
v1uses the legacy/v1/chatendpoint;v2(default) uses/v2/chatand supports tool calling. possible known values include one of ["v1", "v2"]; Default: "v2" - cache_
read_ numbercost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ numbercost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ list(object)cost_ lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ list(object)factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embedding_
input_ stringtype - The intended downstream use of the embeddings to improve model quality. possible known values include one of ["classification", "clustering", "image", "searchdocument", "searchquery"]; Default: "classification"
- embeddings_
dimensions number - The number of dimensions for embedding outputs.
- input_
cost number - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens number - The maximum number of tokens to generate in the response.
- output_
cost number - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ list(object)factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k number - Limits the number of highest-probability tokens considered during generation.
- top_
p number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url string - The upstream URL for the model endpoint.
- wait_
for_ boolmodel - Whether to wait for the model to be ready before sending the request. Default: false
- api
Version String - Cohere API version.
v1uses the legacy/v1/chatendpoint;v2(default) uses/v2/chatand supports tool calling. possible known values include one of ["v1", "v2"]; Default: "v2" - cache
Read DoubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write DoubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<AiCost Lists Gateway Model Api Target Config Cohere Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<AiFactors Gateway Model Api Target Config Cohere Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embedding
Input StringType - The intended downstream use of the embeddings to improve model quality. possible known values include one of ["classification", "clustering", "image", "searchdocument", "searchquery"]; Default: "classification"
- embeddings
Dimensions Double - The number of dimensions for embedding outputs.
- input
Cost Double - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Double - The maximum number of tokens to generate in the response.
- output
Cost Double - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<AiFactors Gateway Model Api Target Config Cohere Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Double
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Double - Limits the number of highest-probability tokens considered during generation.
- top
P Double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
- wait
For BooleanModel - Whether to wait for the model to be ready before sending the request. Default: false
- api
Version string - Cohere API version.
v1uses the legacy/v1/chatendpoint;v2(default) uses/v2/chatand supports tool calling. possible known values include one of ["v1", "v2"]; Default: "v2" - cache
Read numberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write numberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write AiCost Lists Gateway Model Api Target Config Cohere Cache Write Cost List[] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window AiFactors Gateway Model Api Target Config Cohere Context Window Factor[] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embedding
Input stringType - The intended downstream use of the embeddings to improve model quality. possible known values include one of ["classification", "clustering", "image", "searchdocument", "searchquery"]; Default: "classification"
- embeddings
Dimensions number - The number of dimensions for embedding outputs.
- input
Cost number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens number - The maximum number of tokens to generate in the response.
- output
Cost number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier AiFactors Gateway Model Api Target Config Cohere Service Tier Factor[] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K number - Limits the number of highest-probability tokens considered during generation.
- top
P number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url string - The upstream URL for the model endpoint.
- wait
For booleanModel - Whether to wait for the model to be ready before sending the request. Default: false
- api_
version str - Cohere API version.
v1uses the legacy/v1/chatendpoint;v2(default) uses/v2/chatand supports tool calling. possible known values include one of ["v1", "v2"]; Default: "v2" - cache_
read_ floatcost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ floatcost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ Sequence[Aicost_ lists Gateway Model Api Target Config Cohere Cache Write Cost List] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ Sequence[Aifactors Gateway Model Api Target Config Cohere Context Window Factor] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embedding_
input_ strtype - The intended downstream use of the embeddings to improve model quality. possible known values include one of ["classification", "clustering", "image", "searchdocument", "searchquery"]; Default: "classification"
- embeddings_
dimensions float - The number of dimensions for embedding outputs.
- input_
cost float - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens float - The maximum number of tokens to generate in the response.
- output_
cost float - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ Sequence[Aifactors Gateway Model Api Target Config Cohere Service Tier Factor] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature float
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k float - Limits the number of highest-probability tokens considered during generation.
- top_
p float - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url str - The upstream URL for the model endpoint.
- wait_
for_ boolmodel - Whether to wait for the model to be ready before sending the request. Default: false
- api
Version String - Cohere API version.
v1uses the legacy/v1/chatendpoint;v2(default) uses/v2/chatand supports tool calling. possible known values include one of ["v1", "v2"]; Default: "v2" - cache
Read NumberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write NumberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<Property Map>Cost Lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<Property Map>Factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embedding
Input StringType - The intended downstream use of the embeddings to improve model quality. possible known values include one of ["classification", "clustering", "image", "searchdocument", "searchquery"]; Default: "classification"
- embeddings
Dimensions Number - The number of dimensions for embedding outputs.
- input
Cost Number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Number - The maximum number of tokens to generate in the response.
- output
Cost Number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<Property Map>Factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Number - Limits the number of highest-probability tokens considered during generation.
- top
P Number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
- wait
For BooleanModel - Whether to wait for the model to be ready before sending the request. Default: false
AiGatewayModelApiTargetConfigCohereCacheWriteCostList, AiGatewayModelApiTargetConfigCohereCacheWriteCostListArgs
AiGatewayModelApiTargetConfigCohereContextWindowFactor, AiGatewayModelApiTargetConfigCohereContextWindowFactorArgs
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor double - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor double - Multiplier applied to output pricing above the threshold. Not Null
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor float64 - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor float64 - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor number - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor number - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Double - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Double - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor number - Multiplier applied to output pricing above the threshold. Not Null
- above str
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor float - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor float - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Number - Multiplier applied to output pricing above the threshold. Not Null
AiGatewayModelApiTargetConfigCohereServiceTierFactor, AiGatewayModelApiTargetConfigCohereServiceTierFactorArgs
- Factor double
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- Factor float64
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Double
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor float
- Multiplier applied to the whole request for this service tier. Not Null
- tier str
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Number
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
AiGatewayModelApiTargetConfigDashscope, AiGatewayModelApiTargetConfigDashscopeArgs
- Cache
Read doubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write doubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write List<AiCost Lists Gateway Model Api Target Config Dashscope Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window List<AiFactors Gateway Model Api Target Config Dashscope Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions double - The number of dimensions for embedding outputs.
- Input
Cost double - Cost per 1M input tokens for billing and cost tracking.
- International bool
- Whether to use the international DashScope endpoint. Default: true
- Max
Tokens double - The maximum number of tokens to generate in the response.
- Output
Cost double - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier List<AiFactors Gateway Model Api Target Config Dashscope Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature double
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K double - Limits the number of highest-probability tokens considered during generation.
- Top
P double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- Cache
Read float64Cost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write float64Cost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write []AiCost Lists Gateway Model Api Target Config Dashscope Cache Write Cost List - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window []AiFactors Gateway Model Api Target Config Dashscope Context Window Factor - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions float64 - The number of dimensions for embedding outputs.
- Input
Cost float64 - Cost per 1M input tokens for billing and cost tracking.
- International bool
- Whether to use the international DashScope endpoint. Default: true
- Max
Tokens float64 - The maximum number of tokens to generate in the response.
- Output
Cost float64 - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier []AiFactors Gateway Model Api Target Config Dashscope Service Tier Factor - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature float64
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K float64 - Limits the number of highest-probability tokens considered during generation.
- Top
P float64 - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- cache_
read_ numbercost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ numbercost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ list(object)cost_ lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ list(object)factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions number - The number of dimensions for embedding outputs.
- input_
cost number - Cost per 1M input tokens for billing and cost tracking.
- international bool
- Whether to use the international DashScope endpoint. Default: true
- max_
tokens number - The maximum number of tokens to generate in the response.
- output_
cost number - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ list(object)factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k number - Limits the number of highest-probability tokens considered during generation.
- top_
p number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url string - The upstream URL for the model endpoint.
- cache
Read DoubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write DoubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<AiCost Lists Gateway Model Api Target Config Dashscope Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<AiFactors Gateway Model Api Target Config Dashscope Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Double - The number of dimensions for embedding outputs.
- input
Cost Double - Cost per 1M input tokens for billing and cost tracking.
- international Boolean
- Whether to use the international DashScope endpoint. Default: true
- max
Tokens Double - The maximum number of tokens to generate in the response.
- output
Cost Double - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<AiFactors Gateway Model Api Target Config Dashscope Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Double
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Double - Limits the number of highest-probability tokens considered during generation.
- top
P Double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
- cache
Read numberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write numberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write AiCost Lists Gateway Model Api Target Config Dashscope Cache Write Cost List[] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window AiFactors Gateway Model Api Target Config Dashscope Context Window Factor[] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions number - The number of dimensions for embedding outputs.
- input
Cost number - Cost per 1M input tokens for billing and cost tracking.
- international boolean
- Whether to use the international DashScope endpoint. Default: true
- max
Tokens number - The maximum number of tokens to generate in the response.
- output
Cost number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier AiFactors Gateway Model Api Target Config Dashscope Service Tier Factor[] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K number - Limits the number of highest-probability tokens considered during generation.
- top
P number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url string - The upstream URL for the model endpoint.
- cache_
read_ floatcost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ floatcost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ Sequence[Aicost_ lists Gateway Model Api Target Config Dashscope Cache Write Cost List] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ Sequence[Aifactors Gateway Model Api Target Config Dashscope Context Window Factor] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions float - The number of dimensions for embedding outputs.
- input_
cost float - Cost per 1M input tokens for billing and cost tracking.
- international bool
- Whether to use the international DashScope endpoint. Default: true
- max_
tokens float - The maximum number of tokens to generate in the response.
- output_
cost float - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ Sequence[Aifactors Gateway Model Api Target Config Dashscope Service Tier Factor] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature float
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k float - Limits the number of highest-probability tokens considered during generation.
- top_
p float - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url str - The upstream URL for the model endpoint.
- cache
Read NumberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write NumberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<Property Map>Cost Lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<Property Map>Factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Number - The number of dimensions for embedding outputs.
- input
Cost Number - Cost per 1M input tokens for billing and cost tracking.
- international Boolean
- Whether to use the international DashScope endpoint. Default: true
- max
Tokens Number - The maximum number of tokens to generate in the response.
- output
Cost Number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<Property Map>Factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Number - Limits the number of highest-probability tokens considered during generation.
- top
P Number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
AiGatewayModelApiTargetConfigDashscopeCacheWriteCostList, AiGatewayModelApiTargetConfigDashscopeCacheWriteCostListArgs
AiGatewayModelApiTargetConfigDashscopeContextWindowFactor, AiGatewayModelApiTargetConfigDashscopeContextWindowFactorArgs
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor double - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor double - Multiplier applied to output pricing above the threshold. Not Null
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor float64 - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor float64 - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor number - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor number - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Double - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Double - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor number - Multiplier applied to output pricing above the threshold. Not Null
- above str
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor float - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor float - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Number - Multiplier applied to output pricing above the threshold. Not Null
AiGatewayModelApiTargetConfigDashscopeServiceTierFactor, AiGatewayModelApiTargetConfigDashscopeServiceTierFactorArgs
- Factor double
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- Factor float64
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Double
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor float
- Multiplier applied to the whole request for this service tier. Not Null
- tier str
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Number
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
AiGatewayModelApiTargetConfigDatabricks, AiGatewayModelApiTargetConfigDatabricksArgs
- Cache
Read doubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write doubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write List<AiCost Lists Gateway Model Api Target Config Databricks Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window List<AiFactors Gateway Model Api Target Config Databricks Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions double - The number of dimensions for embedding outputs.
- Input
Cost double - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens double - The maximum number of tokens to generate in the response.
- Output
Cost double - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier List<AiFactors Gateway Model Api Target Config Databricks Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature double
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K double - Limits the number of highest-probability tokens considered during generation.
- Top
P double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- Workspace
Instance stringId - The Databricks workspace instance ID. Not Null
- Cache
Read float64Cost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write float64Cost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write []AiCost Lists Gateway Model Api Target Config Databricks Cache Write Cost List - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window []AiFactors Gateway Model Api Target Config Databricks Context Window Factor - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions float64 - The number of dimensions for embedding outputs.
- Input
Cost float64 - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens float64 - The maximum number of tokens to generate in the response.
- Output
Cost float64 - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier []AiFactors Gateway Model Api Target Config Databricks Service Tier Factor - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature float64
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K float64 - Limits the number of highest-probability tokens considered during generation.
- Top
P float64 - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- Workspace
Instance stringId - The Databricks workspace instance ID. Not Null
- cache_
read_ numbercost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ numbercost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ list(object)cost_ lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ list(object)factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions number - The number of dimensions for embedding outputs.
- input_
cost number - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens number - The maximum number of tokens to generate in the response.
- output_
cost number - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ list(object)factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k number - Limits the number of highest-probability tokens considered during generation.
- top_
p number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url string - The upstream URL for the model endpoint.
- workspace_
instance_ stringid - The Databricks workspace instance ID. Not Null
- cache
Read DoubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write DoubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<AiCost Lists Gateway Model Api Target Config Databricks Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<AiFactors Gateway Model Api Target Config Databricks Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Double - The number of dimensions for embedding outputs.
- input
Cost Double - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Double - The maximum number of tokens to generate in the response.
- output
Cost Double - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<AiFactors Gateway Model Api Target Config Databricks Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Double
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Double - Limits the number of highest-probability tokens considered during generation.
- top
P Double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
- workspace
Instance StringId - The Databricks workspace instance ID. Not Null
- cache
Read numberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write numberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write AiCost Lists Gateway Model Api Target Config Databricks Cache Write Cost List[] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window AiFactors Gateway Model Api Target Config Databricks Context Window Factor[] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions number - The number of dimensions for embedding outputs.
- input
Cost number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens number - The maximum number of tokens to generate in the response.
- output
Cost number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier AiFactors Gateway Model Api Target Config Databricks Service Tier Factor[] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K number - Limits the number of highest-probability tokens considered during generation.
- top
P number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url string - The upstream URL for the model endpoint.
- workspace
Instance stringId - The Databricks workspace instance ID. Not Null
- cache_
read_ floatcost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ floatcost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ Sequence[Aicost_ lists Gateway Model Api Target Config Databricks Cache Write Cost List] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ Sequence[Aifactors Gateway Model Api Target Config Databricks Context Window Factor] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions float - The number of dimensions for embedding outputs.
- input_
cost float - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens float - The maximum number of tokens to generate in the response.
- output_
cost float - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ Sequence[Aifactors Gateway Model Api Target Config Databricks Service Tier Factor] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature float
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k float - Limits the number of highest-probability tokens considered during generation.
- top_
p float - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url str - The upstream URL for the model endpoint.
- workspace_
instance_ strid - The Databricks workspace instance ID. Not Null
- cache
Read NumberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write NumberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<Property Map>Cost Lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<Property Map>Factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Number - The number of dimensions for embedding outputs.
- input
Cost Number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Number - The maximum number of tokens to generate in the response.
- output
Cost Number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<Property Map>Factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Number - Limits the number of highest-probability tokens considered during generation.
- top
P Number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
- workspace
Instance StringId - The Databricks workspace instance ID. Not Null
AiGatewayModelApiTargetConfigDatabricksCacheWriteCostList, AiGatewayModelApiTargetConfigDatabricksCacheWriteCostListArgs
AiGatewayModelApiTargetConfigDatabricksContextWindowFactor, AiGatewayModelApiTargetConfigDatabricksContextWindowFactorArgs
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor double - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor double - Multiplier applied to output pricing above the threshold. Not Null
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor float64 - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor float64 - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor number - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor number - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Double - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Double - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor number - Multiplier applied to output pricing above the threshold. Not Null
- above str
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor float - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor float - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Number - Multiplier applied to output pricing above the threshold. Not Null
AiGatewayModelApiTargetConfigDatabricksServiceTierFactor, AiGatewayModelApiTargetConfigDatabricksServiceTierFactorArgs
- Factor double
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- Factor float64
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Double
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor float
- Multiplier applied to the whole request for this service tier. Not Null
- tier str
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Number
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
AiGatewayModelApiTargetConfigDeepseek, AiGatewayModelApiTargetConfigDeepseekArgs
- Cache
Read doubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write doubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write List<AiCost Lists Gateway Model Api Target Config Deepseek Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window List<AiFactors Gateway Model Api Target Config Deepseek Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions double - The number of dimensions for embedding outputs.
- Input
Cost double - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens double - The maximum number of tokens to generate in the response.
- Output
Cost double - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier List<AiFactors Gateway Model Api Target Config Deepseek Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature double
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K double - Limits the number of highest-probability tokens considered during generation.
- Top
P double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- Cache
Read float64Cost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write float64Cost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write []AiCost Lists Gateway Model Api Target Config Deepseek Cache Write Cost List - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window []AiFactors Gateway Model Api Target Config Deepseek Context Window Factor - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions float64 - The number of dimensions for embedding outputs.
- Input
Cost float64 - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens float64 - The maximum number of tokens to generate in the response.
- Output
Cost float64 - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier []AiFactors Gateway Model Api Target Config Deepseek Service Tier Factor - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature float64
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K float64 - Limits the number of highest-probability tokens considered during generation.
- Top
P float64 - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- cache_
read_ numbercost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ numbercost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ list(object)cost_ lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ list(object)factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions number - The number of dimensions for embedding outputs.
- input_
cost number - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens number - The maximum number of tokens to generate in the response.
- output_
cost number - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ list(object)factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k number - Limits the number of highest-probability tokens considered during generation.
- top_
p number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url string - The upstream URL for the model endpoint.
- cache
Read DoubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write DoubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<AiCost Lists Gateway Model Api Target Config Deepseek Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<AiFactors Gateway Model Api Target Config Deepseek Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Double - The number of dimensions for embedding outputs.
- input
Cost Double - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Double - The maximum number of tokens to generate in the response.
- output
Cost Double - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<AiFactors Gateway Model Api Target Config Deepseek Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Double
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Double - Limits the number of highest-probability tokens considered during generation.
- top
P Double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
- cache
Read numberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write numberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write AiCost Lists Gateway Model Api Target Config Deepseek Cache Write Cost List[] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window AiFactors Gateway Model Api Target Config Deepseek Context Window Factor[] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions number - The number of dimensions for embedding outputs.
- input
Cost number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens number - The maximum number of tokens to generate in the response.
- output
Cost number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier AiFactors Gateway Model Api Target Config Deepseek Service Tier Factor[] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K number - Limits the number of highest-probability tokens considered during generation.
- top
P number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url string - The upstream URL for the model endpoint.
- cache_
read_ floatcost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ floatcost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ Sequence[Aicost_ lists Gateway Model Api Target Config Deepseek Cache Write Cost List] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ Sequence[Aifactors Gateway Model Api Target Config Deepseek Context Window Factor] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions float - The number of dimensions for embedding outputs.
- input_
cost float - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens float - The maximum number of tokens to generate in the response.
- output_
cost float - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ Sequence[Aifactors Gateway Model Api Target Config Deepseek Service Tier Factor] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature float
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k float - Limits the number of highest-probability tokens considered during generation.
- top_
p float - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url str - The upstream URL for the model endpoint.
- cache
Read NumberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write NumberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<Property Map>Cost Lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<Property Map>Factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Number - The number of dimensions for embedding outputs.
- input
Cost Number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Number - The maximum number of tokens to generate in the response.
- output
Cost Number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<Property Map>Factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Number - Limits the number of highest-probability tokens considered during generation.
- top
P Number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
AiGatewayModelApiTargetConfigDeepseekCacheWriteCostList, AiGatewayModelApiTargetConfigDeepseekCacheWriteCostListArgs
AiGatewayModelApiTargetConfigDeepseekContextWindowFactor, AiGatewayModelApiTargetConfigDeepseekContextWindowFactorArgs
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor double - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor double - Multiplier applied to output pricing above the threshold. Not Null
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor float64 - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor float64 - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor number - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor number - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Double - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Double - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor number - Multiplier applied to output pricing above the threshold. Not Null
- above str
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor float - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor float - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Number - Multiplier applied to output pricing above the threshold. Not Null
AiGatewayModelApiTargetConfigDeepseekServiceTierFactor, AiGatewayModelApiTargetConfigDeepseekServiceTierFactorArgs
- Factor double
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- Factor float64
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Double
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor float
- Multiplier applied to the whole request for this service tier. Not Null
- tier str
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Number
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
AiGatewayModelApiTargetConfigGemini, AiGatewayModelApiTargetConfigGeminiArgs
- Cache
Read doubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write doubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write List<AiCost Lists Gateway Model Api Target Config Gemini Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window List<AiFactors Gateway Model Api Target Config Gemini Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions double - The number of dimensions for embedding outputs.
- Gcp
Environment AiGateway Model Api Target Config Gemini Gcp Environment - Configuration for a model hosted on Google Cloud Project.
- Input
Cost double - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens double - The maximum number of tokens to generate in the response.
- Output
Cost double - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier List<AiFactors Gateway Model Api Target Config Gemini Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature double
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K double - Limits the number of highest-probability tokens considered during generation.
- Top
P double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- Cache
Read float64Cost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write float64Cost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write []AiCost Lists Gateway Model Api Target Config Gemini Cache Write Cost List - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window []AiFactors Gateway Model Api Target Config Gemini Context Window Factor - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions float64 - The number of dimensions for embedding outputs.
- Gcp
Environment AiGateway Model Api Target Config Gemini Gcp Environment - Configuration for a model hosted on Google Cloud Project.
- Input
Cost float64 - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens float64 - The maximum number of tokens to generate in the response.
- Output
Cost float64 - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier []AiFactors Gateway Model Api Target Config Gemini Service Tier Factor - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature float64
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K float64 - Limits the number of highest-probability tokens considered during generation.
- Top
P float64 - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- cache_
read_ numbercost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ numbercost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ list(object)cost_ lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ list(object)factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions number - The number of dimensions for embedding outputs.
- gcp_
environment object - Configuration for a model hosted on Google Cloud Project.
- input_
cost number - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens number - The maximum number of tokens to generate in the response.
- output_
cost number - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ list(object)factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k number - Limits the number of highest-probability tokens considered during generation.
- top_
p number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url string - The upstream URL for the model endpoint.
- cache
Read DoubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write DoubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<AiCost Lists Gateway Model Api Target Config Gemini Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<AiFactors Gateway Model Api Target Config Gemini Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Double - The number of dimensions for embedding outputs.
- gcp
Environment AiGateway Model Api Target Config Gemini Gcp Environment - Configuration for a model hosted on Google Cloud Project.
- input
Cost Double - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Double - The maximum number of tokens to generate in the response.
- output
Cost Double - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<AiFactors Gateway Model Api Target Config Gemini Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Double
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Double - Limits the number of highest-probability tokens considered during generation.
- top
P Double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
- cache
Read numberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write numberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write AiCost Lists Gateway Model Api Target Config Gemini Cache Write Cost List[] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window AiFactors Gateway Model Api Target Config Gemini Context Window Factor[] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions number - The number of dimensions for embedding outputs.
- gcp
Environment AiGateway Model Api Target Config Gemini Gcp Environment - Configuration for a model hosted on Google Cloud Project.
- input
Cost number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens number - The maximum number of tokens to generate in the response.
- output
Cost number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier AiFactors Gateway Model Api Target Config Gemini Service Tier Factor[] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K number - Limits the number of highest-probability tokens considered during generation.
- top
P number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url string - The upstream URL for the model endpoint.
- cache_
read_ floatcost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ floatcost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ Sequence[Aicost_ lists Gateway Model Api Target Config Gemini Cache Write Cost List] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ Sequence[Aifactors Gateway Model Api Target Config Gemini Context Window Factor] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions float - The number of dimensions for embedding outputs.
- gcp_
environment AiGateway Model Api Target Config Gemini Gcp Environment - Configuration for a model hosted on Google Cloud Project.
- input_
cost float - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens float - The maximum number of tokens to generate in the response.
- output_
cost float - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ Sequence[Aifactors Gateway Model Api Target Config Gemini Service Tier Factor] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature float
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k float - Limits the number of highest-probability tokens considered during generation.
- top_
p float - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url str - The upstream URL for the model endpoint.
- cache
Read NumberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write NumberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<Property Map>Cost Lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<Property Map>Factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Number - The number of dimensions for embedding outputs.
- gcp
Environment Property Map - Configuration for a model hosted on Google Cloud Project.
- input
Cost Number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Number - The maximum number of tokens to generate in the response.
- output
Cost Number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<Property Map>Factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Number - Limits the number of highest-probability tokens considered during generation.
- top
P Number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
AiGatewayModelApiTargetConfigGeminiCacheWriteCostList, AiGatewayModelApiTargetConfigGeminiCacheWriteCostListArgs
AiGatewayModelApiTargetConfigGeminiContextWindowFactor, AiGatewayModelApiTargetConfigGeminiContextWindowFactorArgs
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor double - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor double - Multiplier applied to output pricing above the threshold. Not Null
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor float64 - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor float64 - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor number - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor number - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Double - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Double - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor number - Multiplier applied to output pricing above the threshold. Not Null
- above str
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor float - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor float - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Number - Multiplier applied to output pricing above the threshold. Not Null
AiGatewayModelApiTargetConfigGeminiGcpEnvironment, AiGatewayModelApiTargetConfigGeminiGcpEnvironmentArgs
- Api
Endpoint string - The custom API endpoint for the Gemini model. Not Null
- Location
Id string - The Google Cloud location ID for the model endpoint. Not Null
- Project
Id string - The Google Cloud project ID for the model endpoint. Not Null
- Api
Endpoint string - The custom API endpoint for the Gemini model. Not Null
- Location
Id string - The Google Cloud location ID for the model endpoint. Not Null
- Project
Id string - The Google Cloud project ID for the model endpoint. Not Null
- api_
endpoint string - The custom API endpoint for the Gemini model. Not Null
- location_
id string - The Google Cloud location ID for the model endpoint. Not Null
- project_
id string - The Google Cloud project ID for the model endpoint. Not Null
- api
Endpoint String - The custom API endpoint for the Gemini model. Not Null
- location
Id String - The Google Cloud location ID for the model endpoint. Not Null
- project
Id String - The Google Cloud project ID for the model endpoint. Not Null
- api
Endpoint string - The custom API endpoint for the Gemini model. Not Null
- location
Id string - The Google Cloud location ID for the model endpoint. Not Null
- project
Id string - The Google Cloud project ID for the model endpoint. Not Null
- api_
endpoint str - The custom API endpoint for the Gemini model. Not Null
- location_
id str - The Google Cloud location ID for the model endpoint. Not Null
- project_
id str - The Google Cloud project ID for the model endpoint. Not Null
- api
Endpoint String - The custom API endpoint for the Gemini model. Not Null
- location
Id String - The Google Cloud location ID for the model endpoint. Not Null
- project
Id String - The Google Cloud project ID for the model endpoint. Not Null
AiGatewayModelApiTargetConfigGeminiServiceTierFactor, AiGatewayModelApiTargetConfigGeminiServiceTierFactorArgs
- Factor double
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- Factor float64
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Double
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor float
- Multiplier applied to the whole request for this service tier. Not Null
- tier str
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Number
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
AiGatewayModelApiTargetConfigHuggingface, AiGatewayModelApiTargetConfigHuggingfaceArgs
- Cache
Read doubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write doubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write List<AiCost Lists Gateway Model Api Target Config Huggingface Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window List<AiFactors Gateway Model Api Target Config Huggingface Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions double - The number of dimensions for embedding outputs.
- Input
Cost double - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens double - The maximum number of tokens to generate in the response.
- Output
Cost double - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier List<AiFactors Gateway Model Api Target Config Huggingface Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature double
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K double - Limits the number of highest-probability tokens considered during generation.
- Top
P double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- Use
Cache bool - Whether to use the Hugging Face inference cache. Default: false
- Wait
For boolModel - Whether to wait for the model to load if it is not ready. Default: false
- Cache
Read float64Cost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write float64Cost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write []AiCost Lists Gateway Model Api Target Config Huggingface Cache Write Cost List - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window []AiFactors Gateway Model Api Target Config Huggingface Context Window Factor - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions float64 - The number of dimensions for embedding outputs.
- Input
Cost float64 - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens float64 - The maximum number of tokens to generate in the response.
- Output
Cost float64 - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier []AiFactors Gateway Model Api Target Config Huggingface Service Tier Factor - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature float64
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K float64 - Limits the number of highest-probability tokens considered during generation.
- Top
P float64 - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- Use
Cache bool - Whether to use the Hugging Face inference cache. Default: false
- Wait
For boolModel - Whether to wait for the model to load if it is not ready. Default: false
- cache_
read_ numbercost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ numbercost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ list(object)cost_ lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ list(object)factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions number - The number of dimensions for embedding outputs.
- input_
cost number - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens number - The maximum number of tokens to generate in the response.
- output_
cost number - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ list(object)factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k number - Limits the number of highest-probability tokens considered during generation.
- top_
p number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url string - The upstream URL for the model endpoint.
- use_
cache bool - Whether to use the Hugging Face inference cache. Default: false
- wait_
for_ boolmodel - Whether to wait for the model to load if it is not ready. Default: false
- cache
Read DoubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write DoubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<AiCost Lists Gateway Model Api Target Config Huggingface Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<AiFactors Gateway Model Api Target Config Huggingface Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Double - The number of dimensions for embedding outputs.
- input
Cost Double - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Double - The maximum number of tokens to generate in the response.
- output
Cost Double - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<AiFactors Gateway Model Api Target Config Huggingface Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Double
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Double - Limits the number of highest-probability tokens considered during generation.
- top
P Double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
- use
Cache Boolean - Whether to use the Hugging Face inference cache. Default: false
- wait
For BooleanModel - Whether to wait for the model to load if it is not ready. Default: false
- cache
Read numberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write numberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write AiCost Lists Gateway Model Api Target Config Huggingface Cache Write Cost List[] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window AiFactors Gateway Model Api Target Config Huggingface Context Window Factor[] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions number - The number of dimensions for embedding outputs.
- input
Cost number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens number - The maximum number of tokens to generate in the response.
- output
Cost number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier AiFactors Gateway Model Api Target Config Huggingface Service Tier Factor[] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K number - Limits the number of highest-probability tokens considered during generation.
- top
P number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url string - The upstream URL for the model endpoint.
- use
Cache boolean - Whether to use the Hugging Face inference cache. Default: false
- wait
For booleanModel - Whether to wait for the model to load if it is not ready. Default: false
- cache_
read_ floatcost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ floatcost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ Sequence[Aicost_ lists Gateway Model Api Target Config Huggingface Cache Write Cost List] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ Sequence[Aifactors Gateway Model Api Target Config Huggingface Context Window Factor] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions float - The number of dimensions for embedding outputs.
- input_
cost float - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens float - The maximum number of tokens to generate in the response.
- output_
cost float - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ Sequence[Aifactors Gateway Model Api Target Config Huggingface Service Tier Factor] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature float
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k float - Limits the number of highest-probability tokens considered during generation.
- top_
p float - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url str - The upstream URL for the model endpoint.
- use_
cache bool - Whether to use the Hugging Face inference cache. Default: false
- wait_
for_ boolmodel - Whether to wait for the model to load if it is not ready. Default: false
- cache
Read NumberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write NumberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<Property Map>Cost Lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<Property Map>Factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Number - The number of dimensions for embedding outputs.
- input
Cost Number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Number - The maximum number of tokens to generate in the response.
- output
Cost Number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<Property Map>Factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Number - Limits the number of highest-probability tokens considered during generation.
- top
P Number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
- use
Cache Boolean - Whether to use the Hugging Face inference cache. Default: false
- wait
For BooleanModel - Whether to wait for the model to load if it is not ready. Default: false
AiGatewayModelApiTargetConfigHuggingfaceCacheWriteCostList, AiGatewayModelApiTargetConfigHuggingfaceCacheWriteCostListArgs
AiGatewayModelApiTargetConfigHuggingfaceContextWindowFactor, AiGatewayModelApiTargetConfigHuggingfaceContextWindowFactorArgs
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor double - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor double - Multiplier applied to output pricing above the threshold. Not Null
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor float64 - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor float64 - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor number - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor number - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Double - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Double - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor number - Multiplier applied to output pricing above the threshold. Not Null
- above str
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor float - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor float - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Number - Multiplier applied to output pricing above the threshold. Not Null
AiGatewayModelApiTargetConfigHuggingfaceServiceTierFactor, AiGatewayModelApiTargetConfigHuggingfaceServiceTierFactorArgs
- Factor double
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- Factor float64
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Double
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor float
- Multiplier applied to the whole request for this service tier. Not Null
- tier str
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Number
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
AiGatewayModelApiTargetConfigKimi, AiGatewayModelApiTargetConfigKimiArgs
- Cache
Read doubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write doubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write List<AiCost Lists Gateway Model Api Target Config Kimi Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window List<AiFactors Gateway Model Api Target Config Kimi Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions double - The number of dimensions for embedding outputs.
- Input
Cost double - Cost per 1M input tokens for billing and cost tracking.
- International bool
- When
true, requests are sent toapi.moonshot.ai(international). Whenfalse, requests are sent toapi.moonshot.cn(mainland China). Default: true - Max
Tokens double - The maximum number of tokens to generate in the response.
- Output
Cost double - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier List<AiFactors Gateway Model Api Target Config Kimi Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature double
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K double - Limits the number of highest-probability tokens considered during generation.
- Top
P double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- Cache
Read float64Cost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write float64Cost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write []AiCost Lists Gateway Model Api Target Config Kimi Cache Write Cost List - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window []AiFactors Gateway Model Api Target Config Kimi Context Window Factor - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions float64 - The number of dimensions for embedding outputs.
- Input
Cost float64 - Cost per 1M input tokens for billing and cost tracking.
- International bool
- When
true, requests are sent toapi.moonshot.ai(international). Whenfalse, requests are sent toapi.moonshot.cn(mainland China). Default: true - Max
Tokens float64 - The maximum number of tokens to generate in the response.
- Output
Cost float64 - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier []AiFactors Gateway Model Api Target Config Kimi Service Tier Factor - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature float64
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K float64 - Limits the number of highest-probability tokens considered during generation.
- Top
P float64 - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- cache_
read_ numbercost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ numbercost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ list(object)cost_ lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ list(object)factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions number - The number of dimensions for embedding outputs.
- input_
cost number - Cost per 1M input tokens for billing and cost tracking.
- international bool
- When
true, requests are sent toapi.moonshot.ai(international). Whenfalse, requests are sent toapi.moonshot.cn(mainland China). Default: true - max_
tokens number - The maximum number of tokens to generate in the response.
- output_
cost number - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ list(object)factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k number - Limits the number of highest-probability tokens considered during generation.
- top_
p number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url string - The upstream URL for the model endpoint.
- cache
Read DoubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write DoubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<AiCost Lists Gateway Model Api Target Config Kimi Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<AiFactors Gateway Model Api Target Config Kimi Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Double - The number of dimensions for embedding outputs.
- input
Cost Double - Cost per 1M input tokens for billing and cost tracking.
- international Boolean
- When
true, requests are sent toapi.moonshot.ai(international). Whenfalse, requests are sent toapi.moonshot.cn(mainland China). Default: true - max
Tokens Double - The maximum number of tokens to generate in the response.
- output
Cost Double - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<AiFactors Gateway Model Api Target Config Kimi Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Double
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Double - Limits the number of highest-probability tokens considered during generation.
- top
P Double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
- cache
Read numberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write numberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write AiCost Lists Gateway Model Api Target Config Kimi Cache Write Cost List[] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window AiFactors Gateway Model Api Target Config Kimi Context Window Factor[] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions number - The number of dimensions for embedding outputs.
- input
Cost number - Cost per 1M input tokens for billing and cost tracking.
- international boolean
- When
true, requests are sent toapi.moonshot.ai(international). Whenfalse, requests are sent toapi.moonshot.cn(mainland China). Default: true - max
Tokens number - The maximum number of tokens to generate in the response.
- output
Cost number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier AiFactors Gateway Model Api Target Config Kimi Service Tier Factor[] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K number - Limits the number of highest-probability tokens considered during generation.
- top
P number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url string - The upstream URL for the model endpoint.
- cache_
read_ floatcost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ floatcost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ Sequence[Aicost_ lists Gateway Model Api Target Config Kimi Cache Write Cost List] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ Sequence[Aifactors Gateway Model Api Target Config Kimi Context Window Factor] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions float - The number of dimensions for embedding outputs.
- input_
cost float - Cost per 1M input tokens for billing and cost tracking.
- international bool
- When
true, requests are sent toapi.moonshot.ai(international). Whenfalse, requests are sent toapi.moonshot.cn(mainland China). Default: true - max_
tokens float - The maximum number of tokens to generate in the response.
- output_
cost float - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ Sequence[Aifactors Gateway Model Api Target Config Kimi Service Tier Factor] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature float
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k float - Limits the number of highest-probability tokens considered during generation.
- top_
p float - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url str - The upstream URL for the model endpoint.
- cache
Read NumberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write NumberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<Property Map>Cost Lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<Property Map>Factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Number - The number of dimensions for embedding outputs.
- input
Cost Number - Cost per 1M input tokens for billing and cost tracking.
- international Boolean
- When
true, requests are sent toapi.moonshot.ai(international). Whenfalse, requests are sent toapi.moonshot.cn(mainland China). Default: true - max
Tokens Number - The maximum number of tokens to generate in the response.
- output
Cost Number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<Property Map>Factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Number - Limits the number of highest-probability tokens considered during generation.
- top
P Number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
AiGatewayModelApiTargetConfigKimiCacheWriteCostList, AiGatewayModelApiTargetConfigKimiCacheWriteCostListArgs
AiGatewayModelApiTargetConfigKimiContextWindowFactor, AiGatewayModelApiTargetConfigKimiContextWindowFactorArgs
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor double - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor double - Multiplier applied to output pricing above the threshold. Not Null
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor float64 - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor float64 - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor number - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor number - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Double - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Double - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor number - Multiplier applied to output pricing above the threshold. Not Null
- above str
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor float - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor float - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Number - Multiplier applied to output pricing above the threshold. Not Null
AiGatewayModelApiTargetConfigKimiServiceTierFactor, AiGatewayModelApiTargetConfigKimiServiceTierFactorArgs
- Factor double
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- Factor float64
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Double
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor float
- Multiplier applied to the whole request for this service tier. Not Null
- tier str
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Number
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
AiGatewayModelApiTargetConfigLlama2, AiGatewayModelApiTargetConfigLlama2Args
- Cache
Read doubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write doubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write List<AiCost Lists Gateway Model Api Target Config Llama2Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window List<AiFactors Gateway Model Api Target Config Llama2Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions double - The number of dimensions for embedding outputs.
- Format string
- The request format to use when communicating with the Llama2 model. possible known values include one of ["ollama", "openai", "raw"]; Not Null
- Input
Cost double - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens double - The maximum number of tokens to generate in the response.
- Output
Cost double - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier List<AiFactors Gateway Model Api Target Config Llama2Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature double
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K double - Limits the number of highest-probability tokens considered during generation.
- Top
P double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint. Not Null
- Cache
Read float64Cost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write float64Cost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write []AiCost Lists Gateway Model Api Target Config Llama2Cache Write Cost List - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window []AiFactors Gateway Model Api Target Config Llama2Context Window Factor - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions float64 - The number of dimensions for embedding outputs.
- Format string
- The request format to use when communicating with the Llama2 model. possible known values include one of ["ollama", "openai", "raw"]; Not Null
- Input
Cost float64 - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens float64 - The maximum number of tokens to generate in the response.
- Output
Cost float64 - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier []AiFactors Gateway Model Api Target Config Llama2Service Tier Factor - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature float64
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K float64 - Limits the number of highest-probability tokens considered during generation.
- Top
P float64 - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint. Not Null
- cache_
read_ numbercost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ numbercost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ list(object)cost_ lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ list(object)factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions number - The number of dimensions for embedding outputs.
- format string
- The request format to use when communicating with the Llama2 model. possible known values include one of ["ollama", "openai", "raw"]; Not Null
- input_
cost number - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens number - The maximum number of tokens to generate in the response.
- output_
cost number - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ list(object)factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k number - Limits the number of highest-probability tokens considered during generation.
- top_
p number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url string - The upstream URL for the model endpoint. Not Null
- cache
Read DoubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write DoubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<AiCost Lists Gateway Model Api Target Config Llama2Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<AiFactors Gateway Model Api Target Config Llama2Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Double - The number of dimensions for embedding outputs.
- format String
- The request format to use when communicating with the Llama2 model. possible known values include one of ["ollama", "openai", "raw"]; Not Null
- input
Cost Double - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Double - The maximum number of tokens to generate in the response.
- output
Cost Double - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<AiFactors Gateway Model Api Target Config Llama2Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Double
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Double - Limits the number of highest-probability tokens considered during generation.
- top
P Double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint. Not Null
- cache
Read numberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write numberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write AiCost Lists Gateway Model Api Target Config Llama2Cache Write Cost List[] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window AiFactors Gateway Model Api Target Config Llama2Context Window Factor[] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions number - The number of dimensions for embedding outputs.
- format string
- The request format to use when communicating with the Llama2 model. possible known values include one of ["ollama", "openai", "raw"]; Not Null
- input
Cost number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens number - The maximum number of tokens to generate in the response.
- output
Cost number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier AiFactors Gateway Model Api Target Config Llama2Service Tier Factor[] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K number - Limits the number of highest-probability tokens considered during generation.
- top
P number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url string - The upstream URL for the model endpoint. Not Null
- cache_
read_ floatcost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ floatcost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ Sequence[Aicost_ lists Gateway Model Api Target Config Llama2Cache Write Cost List] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ Sequence[Aifactors Gateway Model Api Target Config Llama2Context Window Factor] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions float - The number of dimensions for embedding outputs.
- format str
- The request format to use when communicating with the Llama2 model. possible known values include one of ["ollama", "openai", "raw"]; Not Null
- input_
cost float - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens float - The maximum number of tokens to generate in the response.
- output_
cost float - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ Sequence[Aifactors Gateway Model Api Target Config Llama2Service Tier Factor] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature float
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k float - Limits the number of highest-probability tokens considered during generation.
- top_
p float - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url str - The upstream URL for the model endpoint. Not Null
- cache
Read NumberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write NumberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<Property Map>Cost Lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<Property Map>Factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Number - The number of dimensions for embedding outputs.
- format String
- The request format to use when communicating with the Llama2 model. possible known values include one of ["ollama", "openai", "raw"]; Not Null
- input
Cost Number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Number - The maximum number of tokens to generate in the response.
- output
Cost Number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<Property Map>Factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Number - Limits the number of highest-probability tokens considered during generation.
- top
P Number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint. Not Null
AiGatewayModelApiTargetConfigLlama2CacheWriteCostList, AiGatewayModelApiTargetConfigLlama2CacheWriteCostListArgs
AiGatewayModelApiTargetConfigLlama2ContextWindowFactor, AiGatewayModelApiTargetConfigLlama2ContextWindowFactorArgs
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor double - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor double - Multiplier applied to output pricing above the threshold. Not Null
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor float64 - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor float64 - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor number - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor number - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Double - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Double - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor number - Multiplier applied to output pricing above the threshold. Not Null
- above str
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor float - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor float - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Number - Multiplier applied to output pricing above the threshold. Not Null
AiGatewayModelApiTargetConfigLlama2ServiceTierFactor, AiGatewayModelApiTargetConfigLlama2ServiceTierFactorArgs
- Factor double
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- Factor float64
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Double
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor float
- Multiplier applied to the whole request for this service tier. Not Null
- tier str
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Number
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
AiGatewayModelApiTargetConfigMistral, AiGatewayModelApiTargetConfigMistralArgs
- Cache
Read doubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write doubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write List<AiCost Lists Gateway Model Api Target Config Mistral Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window List<AiFactors Gateway Model Api Target Config Mistral Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions double - The number of dimensions for embedding outputs.
- Format string
- The request format to use when communicating with the Mistral model. possible known values include one of ["ollama", "openai"]; Not Null
- Input
Cost double - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens double - The maximum number of tokens to generate in the response.
- Output
Cost double - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier List<AiFactors Gateway Model Api Target Config Mistral Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature double
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K double - Limits the number of highest-probability tokens considered during generation.
- Top
P double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- Cache
Read float64Cost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write float64Cost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write []AiCost Lists Gateway Model Api Target Config Mistral Cache Write Cost List - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window []AiFactors Gateway Model Api Target Config Mistral Context Window Factor - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions float64 - The number of dimensions for embedding outputs.
- Format string
- The request format to use when communicating with the Mistral model. possible known values include one of ["ollama", "openai"]; Not Null
- Input
Cost float64 - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens float64 - The maximum number of tokens to generate in the response.
- Output
Cost float64 - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier []AiFactors Gateway Model Api Target Config Mistral Service Tier Factor - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature float64
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K float64 - Limits the number of highest-probability tokens considered during generation.
- Top
P float64 - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- cache_
read_ numbercost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ numbercost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ list(object)cost_ lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ list(object)factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions number - The number of dimensions for embedding outputs.
- format string
- The request format to use when communicating with the Mistral model. possible known values include one of ["ollama", "openai"]; Not Null
- input_
cost number - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens number - The maximum number of tokens to generate in the response.
- output_
cost number - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ list(object)factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k number - Limits the number of highest-probability tokens considered during generation.
- top_
p number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url string - The upstream URL for the model endpoint.
- cache
Read DoubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write DoubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<AiCost Lists Gateway Model Api Target Config Mistral Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<AiFactors Gateway Model Api Target Config Mistral Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Double - The number of dimensions for embedding outputs.
- format String
- The request format to use when communicating with the Mistral model. possible known values include one of ["ollama", "openai"]; Not Null
- input
Cost Double - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Double - The maximum number of tokens to generate in the response.
- output
Cost Double - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<AiFactors Gateway Model Api Target Config Mistral Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Double
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Double - Limits the number of highest-probability tokens considered during generation.
- top
P Double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
- cache
Read numberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write numberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write AiCost Lists Gateway Model Api Target Config Mistral Cache Write Cost List[] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window AiFactors Gateway Model Api Target Config Mistral Context Window Factor[] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions number - The number of dimensions for embedding outputs.
- format string
- The request format to use when communicating with the Mistral model. possible known values include one of ["ollama", "openai"]; Not Null
- input
Cost number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens number - The maximum number of tokens to generate in the response.
- output
Cost number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier AiFactors Gateway Model Api Target Config Mistral Service Tier Factor[] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K number - Limits the number of highest-probability tokens considered during generation.
- top
P number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url string - The upstream URL for the model endpoint.
- cache_
read_ floatcost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ floatcost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ Sequence[Aicost_ lists Gateway Model Api Target Config Mistral Cache Write Cost List] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ Sequence[Aifactors Gateway Model Api Target Config Mistral Context Window Factor] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions float - The number of dimensions for embedding outputs.
- format str
- The request format to use when communicating with the Mistral model. possible known values include one of ["ollama", "openai"]; Not Null
- input_
cost float - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens float - The maximum number of tokens to generate in the response.
- output_
cost float - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ Sequence[Aifactors Gateway Model Api Target Config Mistral Service Tier Factor] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature float
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k float - Limits the number of highest-probability tokens considered during generation.
- top_
p float - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url str - The upstream URL for the model endpoint.
- cache
Read NumberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write NumberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<Property Map>Cost Lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<Property Map>Factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Number - The number of dimensions for embedding outputs.
- format String
- The request format to use when communicating with the Mistral model. possible known values include one of ["ollama", "openai"]; Not Null
- input
Cost Number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Number - The maximum number of tokens to generate in the response.
- output
Cost Number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<Property Map>Factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Number - Limits the number of highest-probability tokens considered during generation.
- top
P Number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
AiGatewayModelApiTargetConfigMistralCacheWriteCostList, AiGatewayModelApiTargetConfigMistralCacheWriteCostListArgs
AiGatewayModelApiTargetConfigMistralContextWindowFactor, AiGatewayModelApiTargetConfigMistralContextWindowFactorArgs
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor double - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor double - Multiplier applied to output pricing above the threshold. Not Null
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor float64 - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor float64 - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor number - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor number - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Double - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Double - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor number - Multiplier applied to output pricing above the threshold. Not Null
- above str
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor float - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor float - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Number - Multiplier applied to output pricing above the threshold. Not Null
AiGatewayModelApiTargetConfigMistralServiceTierFactor, AiGatewayModelApiTargetConfigMistralServiceTierFactorArgs
- Factor double
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- Factor float64
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Double
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor float
- Multiplier applied to the whole request for this service tier. Not Null
- tier str
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Number
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
AiGatewayModelApiTargetConfigOllama, AiGatewayModelApiTargetConfigOllamaArgs
- Cache
Read doubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write doubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write List<AiCost Lists Gateway Model Api Target Config Ollama Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window List<AiFactors Gateway Model Api Target Config Ollama Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions double - The number of dimensions for embedding outputs.
- Input
Cost double - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens double - The maximum number of tokens to generate in the response.
- Output
Cost double - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier List<AiFactors Gateway Model Api Target Config Ollama Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature double
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K double - Limits the number of highest-probability tokens considered during generation.
- Top
P double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- Cache
Read float64Cost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write float64Cost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write []AiCost Lists Gateway Model Api Target Config Ollama Cache Write Cost List - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window []AiFactors Gateway Model Api Target Config Ollama Context Window Factor - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions float64 - The number of dimensions for embedding outputs.
- Input
Cost float64 - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens float64 - The maximum number of tokens to generate in the response.
- Output
Cost float64 - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier []AiFactors Gateway Model Api Target Config Ollama Service Tier Factor - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature float64
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K float64 - Limits the number of highest-probability tokens considered during generation.
- Top
P float64 - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- cache_
read_ numbercost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ numbercost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ list(object)cost_ lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ list(object)factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions number - The number of dimensions for embedding outputs.
- input_
cost number - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens number - The maximum number of tokens to generate in the response.
- output_
cost number - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ list(object)factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k number - Limits the number of highest-probability tokens considered during generation.
- top_
p number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url string - The upstream URL for the model endpoint.
- cache
Read DoubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write DoubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<AiCost Lists Gateway Model Api Target Config Ollama Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<AiFactors Gateway Model Api Target Config Ollama Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Double - The number of dimensions for embedding outputs.
- input
Cost Double - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Double - The maximum number of tokens to generate in the response.
- output
Cost Double - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<AiFactors Gateway Model Api Target Config Ollama Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Double
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Double - Limits the number of highest-probability tokens considered during generation.
- top
P Double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
- cache
Read numberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write numberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write AiCost Lists Gateway Model Api Target Config Ollama Cache Write Cost List[] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window AiFactors Gateway Model Api Target Config Ollama Context Window Factor[] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions number - The number of dimensions for embedding outputs.
- input
Cost number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens number - The maximum number of tokens to generate in the response.
- output
Cost number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier AiFactors Gateway Model Api Target Config Ollama Service Tier Factor[] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K number - Limits the number of highest-probability tokens considered during generation.
- top
P number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url string - The upstream URL for the model endpoint.
- cache_
read_ floatcost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ floatcost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ Sequence[Aicost_ lists Gateway Model Api Target Config Ollama Cache Write Cost List] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ Sequence[Aifactors Gateway Model Api Target Config Ollama Context Window Factor] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions float - The number of dimensions for embedding outputs.
- input_
cost float - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens float - The maximum number of tokens to generate in the response.
- output_
cost float - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ Sequence[Aifactors Gateway Model Api Target Config Ollama Service Tier Factor] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature float
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k float - Limits the number of highest-probability tokens considered during generation.
- top_
p float - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url str - The upstream URL for the model endpoint.
- cache
Read NumberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write NumberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<Property Map>Cost Lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<Property Map>Factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Number - The number of dimensions for embedding outputs.
- input
Cost Number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Number - The maximum number of tokens to generate in the response.
- output
Cost Number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<Property Map>Factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Number - Limits the number of highest-probability tokens considered during generation.
- top
P Number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
AiGatewayModelApiTargetConfigOllamaCacheWriteCostList, AiGatewayModelApiTargetConfigOllamaCacheWriteCostListArgs
AiGatewayModelApiTargetConfigOllamaContextWindowFactor, AiGatewayModelApiTargetConfigOllamaContextWindowFactorArgs
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor double - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor double - Multiplier applied to output pricing above the threshold. Not Null
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor float64 - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor float64 - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor number - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor number - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Double - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Double - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor number - Multiplier applied to output pricing above the threshold. Not Null
- above str
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor float - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor float - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Number - Multiplier applied to output pricing above the threshold. Not Null
AiGatewayModelApiTargetConfigOllamaServiceTierFactor, AiGatewayModelApiTargetConfigOllamaServiceTierFactorArgs
- Factor double
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- Factor float64
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Double
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor float
- Multiplier applied to the whole request for this service tier. Not Null
- tier str
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Number
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
AiGatewayModelApiTargetConfigOpenai, AiGatewayModelApiTargetConfigOpenaiArgs
- Cache
Read doubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write doubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write List<AiCost Lists Gateway Model Api Target Config Openai Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window List<AiFactors Gateway Model Api Target Config Openai Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions double - The number of dimensions for embedding outputs.
- Input
Cost double - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens double - The maximum number of tokens to generate in the response.
- Output
Cost double - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier List<AiFactors Gateway Model Api Target Config Openai Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature double
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K double - Limits the number of highest-probability tokens considered during generation.
- Top
P double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- Cache
Read float64Cost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write float64Cost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write []AiCost Lists Gateway Model Api Target Config Openai Cache Write Cost List - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window []AiFactors Gateway Model Api Target Config Openai Context Window Factor - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions float64 - The number of dimensions for embedding outputs.
- Input
Cost float64 - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens float64 - The maximum number of tokens to generate in the response.
- Output
Cost float64 - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier []AiFactors Gateway Model Api Target Config Openai Service Tier Factor - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature float64
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K float64 - Limits the number of highest-probability tokens considered during generation.
- Top
P float64 - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- cache_
read_ numbercost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ numbercost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ list(object)cost_ lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ list(object)factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions number - The number of dimensions for embedding outputs.
- input_
cost number - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens number - The maximum number of tokens to generate in the response.
- output_
cost number - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ list(object)factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k number - Limits the number of highest-probability tokens considered during generation.
- top_
p number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url string - The upstream URL for the model endpoint.
- cache
Read DoubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write DoubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<AiCost Lists Gateway Model Api Target Config Openai Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<AiFactors Gateway Model Api Target Config Openai Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Double - The number of dimensions for embedding outputs.
- input
Cost Double - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Double - The maximum number of tokens to generate in the response.
- output
Cost Double - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<AiFactors Gateway Model Api Target Config Openai Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Double
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Double - Limits the number of highest-probability tokens considered during generation.
- top
P Double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
- cache
Read numberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write numberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write AiCost Lists Gateway Model Api Target Config Openai Cache Write Cost List[] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window AiFactors Gateway Model Api Target Config Openai Context Window Factor[] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions number - The number of dimensions for embedding outputs.
- input
Cost number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens number - The maximum number of tokens to generate in the response.
- output
Cost number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier AiFactors Gateway Model Api Target Config Openai Service Tier Factor[] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K number - Limits the number of highest-probability tokens considered during generation.
- top
P number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url string - The upstream URL for the model endpoint.
- cache_
read_ floatcost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ floatcost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ Sequence[Aicost_ lists Gateway Model Api Target Config Openai Cache Write Cost List] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ Sequence[Aifactors Gateway Model Api Target Config Openai Context Window Factor] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions float - The number of dimensions for embedding outputs.
- input_
cost float - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens float - The maximum number of tokens to generate in the response.
- output_
cost float - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ Sequence[Aifactors Gateway Model Api Target Config Openai Service Tier Factor] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature float
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k float - Limits the number of highest-probability tokens considered during generation.
- top_
p float - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url str - The upstream URL for the model endpoint.
- cache
Read NumberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write NumberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<Property Map>Cost Lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<Property Map>Factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Number - The number of dimensions for embedding outputs.
- input
Cost Number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Number - The maximum number of tokens to generate in the response.
- output
Cost Number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<Property Map>Factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Number - Limits the number of highest-probability tokens considered during generation.
- top
P Number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
AiGatewayModelApiTargetConfigOpenaiCacheWriteCostList, AiGatewayModelApiTargetConfigOpenaiCacheWriteCostListArgs
AiGatewayModelApiTargetConfigOpenaiContextWindowFactor, AiGatewayModelApiTargetConfigOpenaiContextWindowFactorArgs
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor double - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor double - Multiplier applied to output pricing above the threshold. Not Null
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor float64 - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor float64 - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor number - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor number - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Double - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Double - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor number - Multiplier applied to output pricing above the threshold. Not Null
- above str
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor float - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor float - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Number - Multiplier applied to output pricing above the threshold. Not Null
AiGatewayModelApiTargetConfigOpenaiServiceTierFactor, AiGatewayModelApiTargetConfigOpenaiServiceTierFactorArgs
- Factor double
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- Factor float64
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Double
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor float
- Multiplier applied to the whole request for this service tier. Not Null
- tier str
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Number
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
AiGatewayModelApiTargetConfigSagemaker, AiGatewayModelApiTargetConfigSagemakerArgs
- Aws
Ai
Gateway Model Api Target Config Sagemaker Aws - Cache
Read doubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write doubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write List<AiCost Lists Gateway Model Api Target Config Sagemaker Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window List<AiFactors Gateway Model Api Target Config Sagemaker Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions double - The number of dimensions for embedding outputs.
- Input
Cost double - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens double - The maximum number of tokens to generate in the response.
- Output
Cost double - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier List<AiFactors Gateway Model Api Target Config Sagemaker Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Target
Ai
Gateway Model Api Target Config Sagemaker Target - Temperature double
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K double - Limits the number of highest-probability tokens considered during generation.
- Top
P double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- Aws
Ai
Gateway Model Api Target Config Sagemaker Aws - Cache
Read float64Cost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write float64Cost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write []AiCost Lists Gateway Model Api Target Config Sagemaker Cache Write Cost List - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window []AiFactors Gateway Model Api Target Config Sagemaker Context Window Factor - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions float64 - The number of dimensions for embedding outputs.
- Input
Cost float64 - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens float64 - The maximum number of tokens to generate in the response.
- Output
Cost float64 - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier []AiFactors Gateway Model Api Target Config Sagemaker Service Tier Factor - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Target
Ai
Gateway Model Api Target Config Sagemaker Target - Temperature float64
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K float64 - Limits the number of highest-probability tokens considered during generation.
- Top
P float64 - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- aws object
- cache_
read_ numbercost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ numbercost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ list(object)cost_ lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ list(object)factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions number - The number of dimensions for embedding outputs.
- input_
cost number - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens number - The maximum number of tokens to generate in the response.
- output_
cost number - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ list(object)factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- target object
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k number - Limits the number of highest-probability tokens considered during generation.
- top_
p number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url string - The upstream URL for the model endpoint.
- aws
Ai
Gateway Model Api Target Config Sagemaker Aws - cache
Read DoubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write DoubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<AiCost Lists Gateway Model Api Target Config Sagemaker Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<AiFactors Gateway Model Api Target Config Sagemaker Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Double - The number of dimensions for embedding outputs.
- input
Cost Double - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Double - The maximum number of tokens to generate in the response.
- output
Cost Double - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<AiFactors Gateway Model Api Target Config Sagemaker Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- target
Ai
Gateway Model Api Target Config Sagemaker Target - temperature Double
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Double - Limits the number of highest-probability tokens considered during generation.
- top
P Double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
- aws
Ai
Gateway Model Api Target Config Sagemaker Aws - cache
Read numberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write numberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write AiCost Lists Gateway Model Api Target Config Sagemaker Cache Write Cost List[] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window AiFactors Gateway Model Api Target Config Sagemaker Context Window Factor[] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions number - The number of dimensions for embedding outputs.
- input
Cost number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens number - The maximum number of tokens to generate in the response.
- output
Cost number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier AiFactors Gateway Model Api Target Config Sagemaker Service Tier Factor[] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- target
Ai
Gateway Model Api Target Config Sagemaker Target - temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K number - Limits the number of highest-probability tokens considered during generation.
- top
P number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url string - The upstream URL for the model endpoint.
- aws
Ai
Gateway Model Api Target Config Sagemaker Aws - cache_
read_ floatcost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ floatcost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ Sequence[Aicost_ lists Gateway Model Api Target Config Sagemaker Cache Write Cost List] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ Sequence[Aifactors Gateway Model Api Target Config Sagemaker Context Window Factor] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions float - The number of dimensions for embedding outputs.
- input_
cost float - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens float - The maximum number of tokens to generate in the response.
- output_
cost float - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ Sequence[Aifactors Gateway Model Api Target Config Sagemaker Service Tier Factor] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- target
Ai
Gateway Model Api Target Config Sagemaker Target - temperature float
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k float - Limits the number of highest-probability tokens considered during generation.
- top_
p float - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url str - The upstream URL for the model endpoint.
- aws Property Map
- cache
Read NumberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write NumberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<Property Map>Cost Lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<Property Map>Factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Number - The number of dimensions for embedding outputs.
- input
Cost Number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Number - The maximum number of tokens to generate in the response.
- output
Cost Number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<Property Map>Factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- target Property Map
- temperature Number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Number - Limits the number of highest-probability tokens considered during generation.
- top
P Number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
AiGatewayModelApiTargetConfigSagemakerAws, AiGatewayModelApiTargetConfigSagemakerAwsArgs
- Assume
Role stringArn - Assume a different IAM role after authenticating; mutually required with rolesessionname.
- Region string
- Overrides the AWS_REGION environment variable for SageMaker requests.
- Role
Session stringName - Session identifier for the assumed role; mutually required with assumerolearn.
- Sts
Endpoint stringUrl - Overrides the STS endpoint when assuming a role.
- Assume
Role stringArn - Assume a different IAM role after authenticating; mutually required with rolesessionname.
- Region string
- Overrides the AWS_REGION environment variable for SageMaker requests.
- Role
Session stringName - Session identifier for the assumed role; mutually required with assumerolearn.
- Sts
Endpoint stringUrl - Overrides the STS endpoint when assuming a role.
- assume_
role_ stringarn - Assume a different IAM role after authenticating; mutually required with rolesessionname.
- region string
- Overrides the AWS_REGION environment variable for SageMaker requests.
- role_
session_ stringname - Session identifier for the assumed role; mutually required with assumerolearn.
- sts_
endpoint_ stringurl - Overrides the STS endpoint when assuming a role.
- assume
Role StringArn - Assume a different IAM role after authenticating; mutually required with rolesessionname.
- region String
- Overrides the AWS_REGION environment variable for SageMaker requests.
- role
Session StringName - Session identifier for the assumed role; mutually required with assumerolearn.
- sts
Endpoint StringUrl - Overrides the STS endpoint when assuming a role.
- assume
Role stringArn - Assume a different IAM role after authenticating; mutually required with rolesessionname.
- region string
- Overrides the AWS_REGION environment variable for SageMaker requests.
- role
Session stringName - Session identifier for the assumed role; mutually required with assumerolearn.
- sts
Endpoint stringUrl - Overrides the STS endpoint when assuming a role.
- assume_
role_ strarn - Assume a different IAM role after authenticating; mutually required with rolesessionname.
- region str
- Overrides the AWS_REGION environment variable for SageMaker requests.
- role_
session_ strname - Session identifier for the assumed role; mutually required with assumerolearn.
- sts_
endpoint_ strurl - Overrides the STS endpoint when assuming a role.
- assume
Role StringArn - Assume a different IAM role after authenticating; mutually required with rolesessionname.
- region String
- Overrides the AWS_REGION environment variable for SageMaker requests.
- role
Session StringName - Session identifier for the assumed role; mutually required with assumerolearn.
- sts
Endpoint StringUrl - Overrides the STS endpoint when assuming a role.
AiGatewayModelApiTargetConfigSagemakerCacheWriteCostList, AiGatewayModelApiTargetConfigSagemakerCacheWriteCostListArgs
AiGatewayModelApiTargetConfigSagemakerContextWindowFactor, AiGatewayModelApiTargetConfigSagemakerContextWindowFactorArgs
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor double - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor double - Multiplier applied to output pricing above the threshold. Not Null
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor float64 - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor float64 - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor number - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor number - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Double - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Double - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor number - Multiplier applied to output pricing above the threshold. Not Null
- above str
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor float - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor float - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Number - Multiplier applied to output pricing above the threshold. Not Null
AiGatewayModelApiTargetConfigSagemakerServiceTierFactor, AiGatewayModelApiTargetConfigSagemakerServiceTierFactorArgs
- Factor double
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- Factor float64
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Double
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor float
- Multiplier applied to the whole request for this service tier. Not Null
- tier str
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Number
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
AiGatewayModelApiTargetConfigSagemakerTarget, AiGatewayModelApiTargetConfigSagemakerTargetArgs
- Container
Hostname string - Sets the X-Amzn-SageMaker-Target-Container-Hostname header (multi-container).
- Model string
- Sets the X-Amzn-SageMaker-Target-Model header (multi-model endpoints).
- Variant string
- Sets the X-Amzn-SageMaker-Target-Variant header (A/B variant testing).
- Container
Hostname string - Sets the X-Amzn-SageMaker-Target-Container-Hostname header (multi-container).
- Model string
- Sets the X-Amzn-SageMaker-Target-Model header (multi-model endpoints).
- Variant string
- Sets the X-Amzn-SageMaker-Target-Variant header (A/B variant testing).
- container_
hostname string - Sets the X-Amzn-SageMaker-Target-Container-Hostname header (multi-container).
- model string
- Sets the X-Amzn-SageMaker-Target-Model header (multi-model endpoints).
- variant string
- Sets the X-Amzn-SageMaker-Target-Variant header (A/B variant testing).
- container
Hostname String - Sets the X-Amzn-SageMaker-Target-Container-Hostname header (multi-container).
- model String
- Sets the X-Amzn-SageMaker-Target-Model header (multi-model endpoints).
- variant String
- Sets the X-Amzn-SageMaker-Target-Variant header (A/B variant testing).
- container
Hostname string - Sets the X-Amzn-SageMaker-Target-Container-Hostname header (multi-container).
- model string
- Sets the X-Amzn-SageMaker-Target-Model header (multi-model endpoints).
- variant string
- Sets the X-Amzn-SageMaker-Target-Variant header (A/B variant testing).
- container_
hostname str - Sets the X-Amzn-SageMaker-Target-Container-Hostname header (multi-container).
- model str
- Sets the X-Amzn-SageMaker-Target-Model header (multi-model endpoints).
- variant str
- Sets the X-Amzn-SageMaker-Target-Variant header (A/B variant testing).
- container
Hostname String - Sets the X-Amzn-SageMaker-Target-Container-Hostname header (multi-container).
- model String
- Sets the X-Amzn-SageMaker-Target-Model header (multi-model endpoints).
- variant String
- Sets the X-Amzn-SageMaker-Target-Variant header (A/B variant testing).
AiGatewayModelApiTargetConfigVercel, AiGatewayModelApiTargetConfigVercelArgs
- Cache
Read doubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write doubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write List<AiCost Lists Gateway Model Api Target Config Vercel Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window List<AiFactors Gateway Model Api Target Config Vercel Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions double - The number of dimensions for embedding outputs.
- Input
Cost double - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens double - The maximum number of tokens to generate in the response.
- Output
Cost double - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier List<AiFactors Gateway Model Api Target Config Vercel Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature double
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K double - Limits the number of highest-probability tokens considered during generation.
- Top
P double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- Cache
Read float64Cost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write float64Cost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write []AiCost Lists Gateway Model Api Target Config Vercel Cache Write Cost List - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window []AiFactors Gateway Model Api Target Config Vercel Context Window Factor - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions float64 - The number of dimensions for embedding outputs.
- Input
Cost float64 - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens float64 - The maximum number of tokens to generate in the response.
- Output
Cost float64 - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier []AiFactors Gateway Model Api Target Config Vercel Service Tier Factor - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature float64
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K float64 - Limits the number of highest-probability tokens considered during generation.
- Top
P float64 - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- cache_
read_ numbercost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ numbercost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ list(object)cost_ lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ list(object)factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions number - The number of dimensions for embedding outputs.
- input_
cost number - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens number - The maximum number of tokens to generate in the response.
- output_
cost number - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ list(object)factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k number - Limits the number of highest-probability tokens considered during generation.
- top_
p number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url string - The upstream URL for the model endpoint.
- cache
Read DoubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write DoubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<AiCost Lists Gateway Model Api Target Config Vercel Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<AiFactors Gateway Model Api Target Config Vercel Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Double - The number of dimensions for embedding outputs.
- input
Cost Double - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Double - The maximum number of tokens to generate in the response.
- output
Cost Double - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<AiFactors Gateway Model Api Target Config Vercel Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Double
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Double - Limits the number of highest-probability tokens considered during generation.
- top
P Double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
- cache
Read numberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write numberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write AiCost Lists Gateway Model Api Target Config Vercel Cache Write Cost List[] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window AiFactors Gateway Model Api Target Config Vercel Context Window Factor[] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions number - The number of dimensions for embedding outputs.
- input
Cost number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens number - The maximum number of tokens to generate in the response.
- output
Cost number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier AiFactors Gateway Model Api Target Config Vercel Service Tier Factor[] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K number - Limits the number of highest-probability tokens considered during generation.
- top
P number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url string - The upstream URL for the model endpoint.
- cache_
read_ floatcost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ floatcost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ Sequence[Aicost_ lists Gateway Model Api Target Config Vercel Cache Write Cost List] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ Sequence[Aifactors Gateway Model Api Target Config Vercel Context Window Factor] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions float - The number of dimensions for embedding outputs.
- input_
cost float - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens float - The maximum number of tokens to generate in the response.
- output_
cost float - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ Sequence[Aifactors Gateway Model Api Target Config Vercel Service Tier Factor] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature float
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k float - Limits the number of highest-probability tokens considered during generation.
- top_
p float - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url str - The upstream URL for the model endpoint.
- cache
Read NumberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write NumberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<Property Map>Cost Lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<Property Map>Factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Number - The number of dimensions for embedding outputs.
- input
Cost Number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Number - The maximum number of tokens to generate in the response.
- output
Cost Number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<Property Map>Factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Number - Limits the number of highest-probability tokens considered during generation.
- top
P Number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
AiGatewayModelApiTargetConfigVercelCacheWriteCostList, AiGatewayModelApiTargetConfigVercelCacheWriteCostListArgs
AiGatewayModelApiTargetConfigVercelContextWindowFactor, AiGatewayModelApiTargetConfigVercelContextWindowFactorArgs
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor double - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor double - Multiplier applied to output pricing above the threshold. Not Null
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor float64 - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor float64 - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor number - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor number - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Double - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Double - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor number - Multiplier applied to output pricing above the threshold. Not Null
- above str
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor float - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor float - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Number - Multiplier applied to output pricing above the threshold. Not Null
AiGatewayModelApiTargetConfigVercelServiceTierFactor, AiGatewayModelApiTargetConfigVercelServiceTierFactorArgs
- Factor double
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- Factor float64
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Double
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor float
- Multiplier applied to the whole request for this service tier. Not Null
- tier str
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Number
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
AiGatewayModelApiTargetConfigVllm, AiGatewayModelApiTargetConfigVllmArgs
- Cache
Read doubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write doubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write List<AiCost Lists Gateway Model Api Target Config Vllm Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window List<AiFactors Gateway Model Api Target Config Vllm Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions double - The number of dimensions for embedding outputs.
- Input
Cost double - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens double - The maximum number of tokens to generate in the response.
- Output
Cost double - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier List<AiFactors Gateway Model Api Target Config Vllm Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature double
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K double - Limits the number of highest-probability tokens considered during generation.
- Top
P double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint. Not Null
- Cache
Read float64Cost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write float64Cost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write []AiCost Lists Gateway Model Api Target Config Vllm Cache Write Cost List - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window []AiFactors Gateway Model Api Target Config Vllm Context Window Factor - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions float64 - The number of dimensions for embedding outputs.
- Input
Cost float64 - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens float64 - The maximum number of tokens to generate in the response.
- Output
Cost float64 - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier []AiFactors Gateway Model Api Target Config Vllm Service Tier Factor - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature float64
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K float64 - Limits the number of highest-probability tokens considered during generation.
- Top
P float64 - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint. Not Null
- cache_
read_ numbercost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ numbercost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ list(object)cost_ lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ list(object)factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions number - The number of dimensions for embedding outputs.
- input_
cost number - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens number - The maximum number of tokens to generate in the response.
- output_
cost number - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ list(object)factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k number - Limits the number of highest-probability tokens considered during generation.
- top_
p number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url string - The upstream URL for the model endpoint. Not Null
- cache
Read DoubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write DoubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<AiCost Lists Gateway Model Api Target Config Vllm Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<AiFactors Gateway Model Api Target Config Vllm Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Double - The number of dimensions for embedding outputs.
- input
Cost Double - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Double - The maximum number of tokens to generate in the response.
- output
Cost Double - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<AiFactors Gateway Model Api Target Config Vllm Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Double
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Double - Limits the number of highest-probability tokens considered during generation.
- top
P Double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint. Not Null
- cache
Read numberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write numberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write AiCost Lists Gateway Model Api Target Config Vllm Cache Write Cost List[] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window AiFactors Gateway Model Api Target Config Vllm Context Window Factor[] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions number - The number of dimensions for embedding outputs.
- input
Cost number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens number - The maximum number of tokens to generate in the response.
- output
Cost number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier AiFactors Gateway Model Api Target Config Vllm Service Tier Factor[] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K number - Limits the number of highest-probability tokens considered during generation.
- top
P number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url string - The upstream URL for the model endpoint. Not Null
- cache_
read_ floatcost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ floatcost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ Sequence[Aicost_ lists Gateway Model Api Target Config Vllm Cache Write Cost List] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ Sequence[Aifactors Gateway Model Api Target Config Vllm Context Window Factor] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions float - The number of dimensions for embedding outputs.
- input_
cost float - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens float - The maximum number of tokens to generate in the response.
- output_
cost float - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ Sequence[Aifactors Gateway Model Api Target Config Vllm Service Tier Factor] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature float
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k float - Limits the number of highest-probability tokens considered during generation.
- top_
p float - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url str - The upstream URL for the model endpoint. Not Null
- cache
Read NumberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write NumberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<Property Map>Cost Lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<Property Map>Factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Number - The number of dimensions for embedding outputs.
- input
Cost Number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Number - The maximum number of tokens to generate in the response.
- output
Cost Number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<Property Map>Factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Number - Limits the number of highest-probability tokens considered during generation.
- top
P Number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint. Not Null
AiGatewayModelApiTargetConfigVllmCacheWriteCostList, AiGatewayModelApiTargetConfigVllmCacheWriteCostListArgs
AiGatewayModelApiTargetConfigVllmContextWindowFactor, AiGatewayModelApiTargetConfigVllmContextWindowFactorArgs
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor double - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor double - Multiplier applied to output pricing above the threshold. Not Null
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor float64 - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor float64 - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor number - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor number - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Double - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Double - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor number - Multiplier applied to output pricing above the threshold. Not Null
- above str
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor float - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor float - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Number - Multiplier applied to output pricing above the threshold. Not Null
AiGatewayModelApiTargetConfigVllmServiceTierFactor, AiGatewayModelApiTargetConfigVllmServiceTierFactorArgs
- Factor double
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- Factor float64
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Double
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor float
- Multiplier applied to the whole request for this service tier. Not Null
- tier str
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Number
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
AiGatewayModelApiTargetConfigXai, AiGatewayModelApiTargetConfigXaiArgs
- Cache
Read doubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write doubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write List<AiCost Lists Gateway Model Api Target Config Xai Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window List<AiFactors Gateway Model Api Target Config Xai Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions double - The number of dimensions for embedding outputs.
- Input
Cost double - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens double - The maximum number of tokens to generate in the response.
- Output
Cost double - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier List<AiFactors Gateway Model Api Target Config Xai Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature double
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K double - Limits the number of highest-probability tokens considered during generation.
- Top
P double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- Cache
Read float64Cost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write float64Cost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write []AiCost Lists Gateway Model Api Target Config Xai Cache Write Cost List - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window []AiFactors Gateway Model Api Target Config Xai Context Window Factor - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions float64 - The number of dimensions for embedding outputs.
- Input
Cost float64 - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens float64 - The maximum number of tokens to generate in the response.
- Output
Cost float64 - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier []AiFactors Gateway Model Api Target Config Xai Service Tier Factor - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature float64
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K float64 - Limits the number of highest-probability tokens considered during generation.
- Top
P float64 - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- cache_
read_ numbercost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ numbercost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ list(object)cost_ lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ list(object)factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions number - The number of dimensions for embedding outputs.
- input_
cost number - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens number - The maximum number of tokens to generate in the response.
- output_
cost number - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ list(object)factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k number - Limits the number of highest-probability tokens considered during generation.
- top_
p number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url string - The upstream URL for the model endpoint.
- cache
Read DoubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write DoubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<AiCost Lists Gateway Model Api Target Config Xai Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<AiFactors Gateway Model Api Target Config Xai Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Double - The number of dimensions for embedding outputs.
- input
Cost Double - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Double - The maximum number of tokens to generate in the response.
- output
Cost Double - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<AiFactors Gateway Model Api Target Config Xai Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Double
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Double - Limits the number of highest-probability tokens considered during generation.
- top
P Double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
- cache
Read numberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write numberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write AiCost Lists Gateway Model Api Target Config Xai Cache Write Cost List[] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window AiFactors Gateway Model Api Target Config Xai Context Window Factor[] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions number - The number of dimensions for embedding outputs.
- input
Cost number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens number - The maximum number of tokens to generate in the response.
- output
Cost number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier AiFactors Gateway Model Api Target Config Xai Service Tier Factor[] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K number - Limits the number of highest-probability tokens considered during generation.
- top
P number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url string - The upstream URL for the model endpoint.
- cache_
read_ floatcost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ floatcost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ Sequence[Aicost_ lists Gateway Model Api Target Config Xai Cache Write Cost List] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ Sequence[Aifactors Gateway Model Api Target Config Xai Context Window Factor] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions float - The number of dimensions for embedding outputs.
- input_
cost float - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens float - The maximum number of tokens to generate in the response.
- output_
cost float - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ Sequence[Aifactors Gateway Model Api Target Config Xai Service Tier Factor] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature float
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k float - Limits the number of highest-probability tokens considered during generation.
- top_
p float - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url str - The upstream URL for the model endpoint.
- cache
Read NumberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write NumberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<Property Map>Cost Lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<Property Map>Factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Number - The number of dimensions for embedding outputs.
- input
Cost Number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Number - The maximum number of tokens to generate in the response.
- output
Cost Number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<Property Map>Factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Number - Limits the number of highest-probability tokens considered during generation.
- top
P Number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
AiGatewayModelApiTargetConfigXaiCacheWriteCostList, AiGatewayModelApiTargetConfigXaiCacheWriteCostListArgs
AiGatewayModelApiTargetConfigXaiContextWindowFactor, AiGatewayModelApiTargetConfigXaiContextWindowFactorArgs
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor double - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor double - Multiplier applied to output pricing above the threshold. Not Null
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor float64 - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor float64 - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor number - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor number - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Double - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Double - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor number - Multiplier applied to output pricing above the threshold. Not Null
- above str
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor float - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor float - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Number - Multiplier applied to output pricing above the threshold. Not Null
AiGatewayModelApiTargetConfigXaiServiceTierFactor, AiGatewayModelApiTargetConfigXaiServiceTierFactorArgs
- Factor double
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- Factor float64
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Double
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor float
- Multiplier applied to the whole request for this service tier. Not Null
- tier str
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Number
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
AiGatewayModelModel, AiGatewayModelModelArgs
- Access
Ai
Gateway Model Model Access - Access control configuration for a model.
- Capabilities List<string>
- List of AI capabilities enabled for this model. Not Null
- Config
Ai
Gateway Model Model Config - Routing, logging, and load balancing configuration for the model. Not Null
- Created
At string - An ISO-8601 timestamp representation of entity creation date.
- Display
Name string - The display name for this model instance. Not Null
- Enabled bool
- Whether the model is enabled. Default: true
- Formats
List<Ai
Gateway Model Model Format> - List of request/response formats supported by this model. Not Null
- Id string
- Contains a unique identifier used for this resource.
- Labels Dictionary<string, string>
- Public labels store information about an entity that can be used for filtering a list of objects.
- Managed
By Dictionary<string, string> - Name string
- A user-defined unique identifier for this model, used as a stable human-readable reference. This value is immutable after creation.
- Policies List<string>
- List of policy references.
- Targets
List<Ai
Gateway Model Model Target> - One or more backend models that this model entry routes to. Not Null
- Updated
At string - An ISO-8601 timestamp representation of entity update date.
- Access
Ai
Gateway Model Model Access - Access control configuration for a model.
- Capabilities []string
- List of AI capabilities enabled for this model. Not Null
- Config
Ai
Gateway Model Model Config - Routing, logging, and load balancing configuration for the model. Not Null
- Created
At string - An ISO-8601 timestamp representation of entity creation date.
- Display
Name string - The display name for this model instance. Not Null
- Enabled bool
- Whether the model is enabled. Default: true
- Formats
[]Ai
Gateway Model Model Format - List of request/response formats supported by this model. Not Null
- Id string
- Contains a unique identifier used for this resource.
- Labels map[string]string
- Public labels store information about an entity that can be used for filtering a list of objects.
- Managed
By map[string]string - Name string
- A user-defined unique identifier for this model, used as a stable human-readable reference. This value is immutable after creation.
- Policies []string
- List of policy references.
- Targets
[]Ai
Gateway Model Model Target - One or more backend models that this model entry routes to. Not Null
- Updated
At string - An ISO-8601 timestamp representation of entity update date.
- access object
- Access control configuration for a model.
- capabilities list(string)
- List of AI capabilities enabled for this model. Not Null
- config object
- Routing, logging, and load balancing configuration for the model. Not Null
- created_
at string - An ISO-8601 timestamp representation of entity creation date.
- display_
name string - The display name for this model instance. Not Null
- enabled bool
- Whether the model is enabled. Default: true
- formats list(object)
- List of request/response formats supported by this model. Not Null
- id string
- Contains a unique identifier used for this resource.
- labels map(string)
- Public labels store information about an entity that can be used for filtering a list of objects.
- managed_
by map(string) - name string
- A user-defined unique identifier for this model, used as a stable human-readable reference. This value is immutable after creation.
- policies list(string)
- List of policy references.
- targets list(object)
- One or more backend models that this model entry routes to. Not Null
- updated_
at string - An ISO-8601 timestamp representation of entity update date.
- access
Ai
Gateway Model Model Access - Access control configuration for a model.
- capabilities List<String>
- List of AI capabilities enabled for this model. Not Null
- config
Ai
Gateway Model Model Config - Routing, logging, and load balancing configuration for the model. Not Null
- created
At String - An ISO-8601 timestamp representation of entity creation date.
- display
Name String - The display name for this model instance. Not Null
- enabled Boolean
- Whether the model is enabled. Default: true
- formats
List<Ai
Gateway Model Model Format> - List of request/response formats supported by this model. Not Null
- id String
- Contains a unique identifier used for this resource.
- labels Map<String,String>
- Public labels store information about an entity that can be used for filtering a list of objects.
- managed
By Map<String,String> - name String
- A user-defined unique identifier for this model, used as a stable human-readable reference. This value is immutable after creation.
- policies List<String>
- List of policy references.
- targets
List<Ai
Gateway Model Model Target> - One or more backend models that this model entry routes to. Not Null
- updated
At String - An ISO-8601 timestamp representation of entity update date.
- access
Ai
Gateway Model Model Access - Access control configuration for a model.
- capabilities string[]
- List of AI capabilities enabled for this model. Not Null
- config
Ai
Gateway Model Model Config - Routing, logging, and load balancing configuration for the model. Not Null
- created
At string - An ISO-8601 timestamp representation of entity creation date.
- display
Name string - The display name for this model instance. Not Null
- enabled boolean
- Whether the model is enabled. Default: true
- formats
Ai
Gateway Model Model Format[] - List of request/response formats supported by this model. Not Null
- id string
- Contains a unique identifier used for this resource.
- labels {[key: string]: string}
- Public labels store information about an entity that can be used for filtering a list of objects.
- managed
By {[key: string]: string} - name string
- A user-defined unique identifier for this model, used as a stable human-readable reference. This value is immutable after creation.
- policies string[]
- List of policy references.
- targets
Ai
Gateway Model Model Target[] - One or more backend models that this model entry routes to. Not Null
- updated
At string - An ISO-8601 timestamp representation of entity update date.
- access
Ai
Gateway Model Model Access - Access control configuration for a model.
- capabilities Sequence[str]
- List of AI capabilities enabled for this model. Not Null
- config
Ai
Gateway Model Model Config - Routing, logging, and load balancing configuration for the model. Not Null
- created_
at str - An ISO-8601 timestamp representation of entity creation date.
- display_
name str - The display name for this model instance. Not Null
- enabled bool
- Whether the model is enabled. Default: true
- formats
Sequence[Ai
Gateway Model Model Format] - List of request/response formats supported by this model. Not Null
- id str
- Contains a unique identifier used for this resource.
- labels Mapping[str, str]
- Public labels store information about an entity that can be used for filtering a list of objects.
- managed_
by Mapping[str, str] - name str
- A user-defined unique identifier for this model, used as a stable human-readable reference. This value is immutable after creation.
- policies Sequence[str]
- List of policy references.
- targets
Sequence[Ai
Gateway Model Model Target] - One or more backend models that this model entry routes to. Not Null
- updated_
at str - An ISO-8601 timestamp representation of entity update date.
- access Property Map
- Access control configuration for a model.
- capabilities List<String>
- List of AI capabilities enabled for this model. Not Null
- config Property Map
- Routing, logging, and load balancing configuration for the model. Not Null
- created
At String - An ISO-8601 timestamp representation of entity creation date.
- display
Name String - The display name for this model instance. Not Null
- enabled Boolean
- Whether the model is enabled. Default: true
- formats List<Property Map>
- List of request/response formats supported by this model. Not Null
- id String
- Contains a unique identifier used for this resource.
- labels Map<String>
- Public labels store information about an entity that can be used for filtering a list of objects.
- managed
By Map<String> - name String
- A user-defined unique identifier for this model, used as a stable human-readable reference. This value is immutable after creation.
- policies List<String>
- List of policy references.
- targets List<Property Map>
- One or more backend models that this model entry routes to. Not Null
- updated
At String - An ISO-8601 timestamp representation of entity update date.
AiGatewayModelModelAccess, AiGatewayModelModelAccessArgs
- Acls
Ai
Gateway Model Model Access Acls - Access control rules. Configure exactly one of
allowordeny. - Auth
Strategies List<string> - List of auth strategies for granting access to the model. At most 1 auth strategy of each auth strategy type can be referenced.
- Identity
Providers List<string> - List of identity providers for granting access to the model. At most 1 identity provider of each identity provider type can be referenced.
- Acls
Ai
Gateway Model Model Access Acls - Access control rules. Configure exactly one of
allowordeny. - Auth
Strategies []string - List of auth strategies for granting access to the model. At most 1 auth strategy of each auth strategy type can be referenced.
- Identity
Providers []string - List of identity providers for granting access to the model. At most 1 identity provider of each identity provider type can be referenced.
- acls object
- Access control rules. Configure exactly one of
allowordeny. - auth_
strategies list(string) - List of auth strategies for granting access to the model. At most 1 auth strategy of each auth strategy type can be referenced.
- identity_
providers list(string) - List of identity providers for granting access to the model. At most 1 identity provider of each identity provider type can be referenced.
- acls
Ai
Gateway Model Model Access Acls - Access control rules. Configure exactly one of
allowordeny. - auth
Strategies List<String> - List of auth strategies for granting access to the model. At most 1 auth strategy of each auth strategy type can be referenced.
- identity
Providers List<String> - List of identity providers for granting access to the model. At most 1 identity provider of each identity provider type can be referenced.
- acls
Ai
Gateway Model Model Access Acls - Access control rules. Configure exactly one of
allowordeny. - auth
Strategies string[] - List of auth strategies for granting access to the model. At most 1 auth strategy of each auth strategy type can be referenced.
- identity
Providers string[] - List of identity providers for granting access to the model. At most 1 identity provider of each identity provider type can be referenced.
- acls
Ai
Gateway Model Model Access Acls - Access control rules. Configure exactly one of
allowordeny. - auth_
strategies Sequence[str] - List of auth strategies for granting access to the model. At most 1 auth strategy of each auth strategy type can be referenced.
- identity_
providers Sequence[str] - List of identity providers for granting access to the model. At most 1 identity provider of each identity provider type can be referenced.
- acls Property Map
- Access control rules. Configure exactly one of
allowordeny. - auth
Strategies List<String> - List of auth strategies for granting access to the model. At most 1 auth strategy of each auth strategy type can be referenced.
- identity
Providers List<String> - List of identity providers for granting access to the model. At most 1 identity provider of each identity provider type can be referenced.
AiGatewayModelModelAccessAcls, AiGatewayModelModelAccessAclsArgs
AiGatewayModelModelConfig, AiGatewayModelModelConfigArgs
- Balancer
Ai
Gateway Model Model Config Balancer - Configuration for a model's load balancer when multiple target models are configured.
- Logging
Ai
Gateway Model Model Config Logging - Configuration for AI Gateway logging.
- Max
Request doubleBody Size - Maximum size of request body to parse. Set to 0 for unlimited. Default: 8388608
- Model
Ai
Gateway Model Model Config Model - Proxy
Ai
Gateway Model Model Config Proxy - HTTP/HTTPS proxy configuration for outbound requests to the upstream AI provider.
- Response
Streaming string - possible known values include one of ["allow", "always", "deny"]; Default: "allow"
- Route
Ai
Gateway Model Model Config Route - Configuration for an AI Gateway route. Not Null
- Balancer
Ai
Gateway Model Model Config Balancer - Configuration for a model's load balancer when multiple target models are configured.
- Logging
Ai
Gateway Model Model Config Logging - Configuration for AI Gateway logging.
- Max
Request float64Body Size - Maximum size of request body to parse. Set to 0 for unlimited. Default: 8388608
- Model
Ai
Gateway Model Model Config Model - Proxy
Ai
Gateway Model Model Config Proxy - HTTP/HTTPS proxy configuration for outbound requests to the upstream AI provider.
- Response
Streaming string - possible known values include one of ["allow", "always", "deny"]; Default: "allow"
- Route
Ai
Gateway Model Model Config Route - Configuration for an AI Gateway route. Not Null
- balancer object
- Configuration for a model's load balancer when multiple target models are configured.
- logging object
- Configuration for AI Gateway logging.
- max_
request_ numberbody_ size - Maximum size of request body to parse. Set to 0 for unlimited. Default: 8388608
- model object
- proxy object
- HTTP/HTTPS proxy configuration for outbound requests to the upstream AI provider.
- response_
streaming string - possible known values include one of ["allow", "always", "deny"]; Default: "allow"
- route object
- Configuration for an AI Gateway route. Not Null
- balancer
Ai
Gateway Model Model Config Balancer - Configuration for a model's load balancer when multiple target models are configured.
- logging
Ai
Gateway Model Model Config Logging - Configuration for AI Gateway logging.
- max
Request DoubleBody Size - Maximum size of request body to parse. Set to 0 for unlimited. Default: 8388608
- model
Ai
Gateway Model Model Config Model - proxy
Ai
Gateway Model Model Config Proxy - HTTP/HTTPS proxy configuration for outbound requests to the upstream AI provider.
- response
Streaming String - possible known values include one of ["allow", "always", "deny"]; Default: "allow"
- route
Ai
Gateway Model Model Config Route - Configuration for an AI Gateway route. Not Null
- balancer
Ai
Gateway Model Model Config Balancer - Configuration for a model's load balancer when multiple target models are configured.
- logging
Ai
Gateway Model Model Config Logging - Configuration for AI Gateway logging.
- max
Request numberBody Size - Maximum size of request body to parse. Set to 0 for unlimited. Default: 8388608
- model
Ai
Gateway Model Model Config Model - proxy
Ai
Gateway Model Model Config Proxy - HTTP/HTTPS proxy configuration for outbound requests to the upstream AI provider.
- response
Streaming string - possible known values include one of ["allow", "always", "deny"]; Default: "allow"
- route
Ai
Gateway Model Model Config Route - Configuration for an AI Gateway route. Not Null
- balancer
Ai
Gateway Model Model Config Balancer - Configuration for a model's load balancer when multiple target models are configured.
- logging
Ai
Gateway Model Model Config Logging - Configuration for AI Gateway logging.
- max_
request_ floatbody_ size - Maximum size of request body to parse. Set to 0 for unlimited. Default: 8388608
- model
Ai
Gateway Model Model Config Model - proxy
Ai
Gateway Model Model Config Proxy - HTTP/HTTPS proxy configuration for outbound requests to the upstream AI provider.
- response_
streaming str - possible known values include one of ["allow", "always", "deny"]; Default: "allow"
- route
Ai
Gateway Model Model Config Route - Configuration for an AI Gateway route. Not Null
- balancer Property Map
- Configuration for a model's load balancer when multiple target models are configured.
- logging Property Map
- Configuration for AI Gateway logging.
- max
Request NumberBody Size - Maximum size of request body to parse. Set to 0 for unlimited. Default: 8388608
- model Property Map
- proxy Property Map
- HTTP/HTTPS proxy configuration for outbound requests to the upstream AI provider.
- response
Streaming String - possible known values include one of ["allow", "always", "deny"]; Default: "allow"
- route Property Map
- Configuration for an AI Gateway route. Not Null
AiGatewayModelModelConfigBalancer, AiGatewayModelModelConfigBalancerArgs
- Consistent
Hashing AiGateway Model Model Config Balancer Consistent Hashing - Least
Connections AiGateway Model Model Config Balancer Least Connections - Lowest
Latency AiGateway Model Model Config Balancer Lowest Latency - Lowest
Usage AiGateway Model Model Config Balancer Lowest Usage - Priority
Ai
Gateway Model Model Config Balancer Priority - Round
Robin AiGateway Model Model Config Balancer Round Robin - Semantic
Ai
Gateway Model Model Config Balancer Semantic
- Consistent
Hashing AiGateway Model Model Config Balancer Consistent Hashing - Least
Connections AiGateway Model Model Config Balancer Least Connections - Lowest
Latency AiGateway Model Model Config Balancer Lowest Latency - Lowest
Usage AiGateway Model Model Config Balancer Lowest Usage - Priority
Ai
Gateway Model Model Config Balancer Priority - Round
Robin AiGateway Model Model Config Balancer Round Robin - Semantic
Ai
Gateway Model Model Config Balancer Semantic
- consistent
Hashing AiGateway Model Model Config Balancer Consistent Hashing - least
Connections AiGateway Model Model Config Balancer Least Connections - lowest
Latency AiGateway Model Model Config Balancer Lowest Latency - lowest
Usage AiGateway Model Model Config Balancer Lowest Usage - priority
Ai
Gateway Model Model Config Balancer Priority - round
Robin AiGateway Model Model Config Balancer Round Robin - semantic
Ai
Gateway Model Model Config Balancer Semantic
- consistent
Hashing AiGateway Model Model Config Balancer Consistent Hashing - least
Connections AiGateway Model Model Config Balancer Least Connections - lowest
Latency AiGateway Model Model Config Balancer Lowest Latency - lowest
Usage AiGateway Model Model Config Balancer Lowest Usage - priority
Ai
Gateway Model Model Config Balancer Priority - round
Robin AiGateway Model Model Config Balancer Round Robin - semantic
Ai
Gateway Model Model Config Balancer Semantic
- consistent_
hashing AiGateway Model Model Config Balancer Consistent Hashing - least_
connections AiGateway Model Model Config Balancer Least Connections - lowest_
latency AiGateway Model Model Config Balancer Lowest Latency - lowest_
usage AiGateway Model Model Config Balancer Lowest Usage - priority
Ai
Gateway Model Model Config Balancer Priority - round_
robin AiGateway Model Model Config Balancer Round Robin - semantic
Ai
Gateway Model Model Config Balancer Semantic
AiGatewayModelModelConfigBalancerConsistentHashing, AiGatewayModelModelConfigBalancerConsistentHashingArgs
- Connect
Timeout double - Default: 60000
- Fail
Timeout double - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - Failover
Criterias List<string> - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- Hash
On stringHeader - The header to use for consistent-hashing. Default: "X-Kong-LLM-Request-ID"
- Max
Fails double - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - Read
Timeout double - Default: 60000
- Retries double
- The number of retries to execute upon failure to proxy. Default: 5
- Slots double
- The number of slots in the load balancer algorithm. Default: 10000
- Write
Timeout double - Default: 60000
- Connect
Timeout float64 - Default: 60000
- Fail
Timeout float64 - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - Failover
Criterias []string - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- Hash
On stringHeader - The header to use for consistent-hashing. Default: "X-Kong-LLM-Request-ID"
- Max
Fails float64 - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - Read
Timeout float64 - Default: 60000
- Retries float64
- The number of retries to execute upon failure to proxy. Default: 5
- Slots float64
- The number of slots in the load balancer algorithm. Default: 10000
- Write
Timeout float64 - Default: 60000
- connect_
timeout number - Default: 60000
- fail_
timeout number - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover_
criterias list(string) - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- hash_
on_ stringheader - The header to use for consistent-hashing. Default: "X-Kong-LLM-Request-ID"
- max_
fails number - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read_
timeout number - Default: 60000
- retries number
- The number of retries to execute upon failure to proxy. Default: 5
- slots number
- The number of slots in the load balancer algorithm. Default: 10000
- write_
timeout number - Default: 60000
- connect
Timeout Double - Default: 60000
- fail
Timeout Double - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover
Criterias List<String> - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- hash
On StringHeader - The header to use for consistent-hashing. Default: "X-Kong-LLM-Request-ID"
- max
Fails Double - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read
Timeout Double - Default: 60000
- retries Double
- The number of retries to execute upon failure to proxy. Default: 5
- slots Double
- The number of slots in the load balancer algorithm. Default: 10000
- write
Timeout Double - Default: 60000
- connect
Timeout number - Default: 60000
- fail
Timeout number - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover
Criterias string[] - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- hash
On stringHeader - The header to use for consistent-hashing. Default: "X-Kong-LLM-Request-ID"
- max
Fails number - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read
Timeout number - Default: 60000
- retries number
- The number of retries to execute upon failure to proxy. Default: 5
- slots number
- The number of slots in the load balancer algorithm. Default: 10000
- write
Timeout number - Default: 60000
- connect_
timeout float - Default: 60000
- fail_
timeout float - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover_
criterias Sequence[str] - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- hash_
on_ strheader - The header to use for consistent-hashing. Default: "X-Kong-LLM-Request-ID"
- max_
fails float - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read_
timeout float - Default: 60000
- retries float
- The number of retries to execute upon failure to proxy. Default: 5
- slots float
- The number of slots in the load balancer algorithm. Default: 10000
- write_
timeout float - Default: 60000
- connect
Timeout Number - Default: 60000
- fail
Timeout Number - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover
Criterias List<String> - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- hash
On StringHeader - The header to use for consistent-hashing. Default: "X-Kong-LLM-Request-ID"
- max
Fails Number - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read
Timeout Number - Default: 60000
- retries Number
- The number of retries to execute upon failure to proxy. Default: 5
- slots Number
- The number of slots in the load balancer algorithm. Default: 10000
- write
Timeout Number - Default: 60000
AiGatewayModelModelConfigBalancerLeastConnections, AiGatewayModelModelConfigBalancerLeastConnectionsArgs
- Connect
Timeout double - Default: 60000
- Fail
Timeout double - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - Failover
Criterias List<string> - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- Max
Fails double - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - Read
Timeout double - Default: 60000
- Retries double
- The number of retries to execute upon failure to proxy. Default: 5
- Slots double
- The number of slots in the load balancer algorithm. Default: 10000
- Write
Timeout double - Default: 60000
- Connect
Timeout float64 - Default: 60000
- Fail
Timeout float64 - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - Failover
Criterias []string - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- Max
Fails float64 - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - Read
Timeout float64 - Default: 60000
- Retries float64
- The number of retries to execute upon failure to proxy. Default: 5
- Slots float64
- The number of slots in the load balancer algorithm. Default: 10000
- Write
Timeout float64 - Default: 60000
- connect_
timeout number - Default: 60000
- fail_
timeout number - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover_
criterias list(string) - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- max_
fails number - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read_
timeout number - Default: 60000
- retries number
- The number of retries to execute upon failure to proxy. Default: 5
- slots number
- The number of slots in the load balancer algorithm. Default: 10000
- write_
timeout number - Default: 60000
- connect
Timeout Double - Default: 60000
- fail
Timeout Double - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover
Criterias List<String> - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- max
Fails Double - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read
Timeout Double - Default: 60000
- retries Double
- The number of retries to execute upon failure to proxy. Default: 5
- slots Double
- The number of slots in the load balancer algorithm. Default: 10000
- write
Timeout Double - Default: 60000
- connect
Timeout number - Default: 60000
- fail
Timeout number - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover
Criterias string[] - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- max
Fails number - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read
Timeout number - Default: 60000
- retries number
- The number of retries to execute upon failure to proxy. Default: 5
- slots number
- The number of slots in the load balancer algorithm. Default: 10000
- write
Timeout number - Default: 60000
- connect_
timeout float - Default: 60000
- fail_
timeout float - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover_
criterias Sequence[str] - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- max_
fails float - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read_
timeout float - Default: 60000
- retries float
- The number of retries to execute upon failure to proxy. Default: 5
- slots float
- The number of slots in the load balancer algorithm. Default: 10000
- write_
timeout float - Default: 60000
- connect
Timeout Number - Default: 60000
- fail
Timeout Number - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover
Criterias List<String> - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- max
Fails Number - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read
Timeout Number - Default: 60000
- retries Number
- The number of retries to execute upon failure to proxy. Default: 5
- slots Number
- The number of slots in the load balancer algorithm. Default: 10000
- write
Timeout Number - Default: 60000
AiGatewayModelModelConfigBalancerLowestLatency, AiGatewayModelModelConfigBalancerLowestLatencyArgs
- Connect
Timeout double - Default: 60000
- Fail
Timeout double - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - Failover
Criterias List<string> - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- Latency
Strategy string - What metrics to use for latency. Available values are:
tpot(time-per-output-token) ande2e. possible known values include one of ["e2e", "tpot"]; Default: "tpot" - Max
Fails double - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - Read
Timeout double - Default: 60000
- Retries double
- The number of retries to execute upon failure to proxy. Default: 5
- Slots double
- The number of slots in the load balancer algorithm. Default: 10000
- Write
Timeout double - Default: 60000
- Connect
Timeout float64 - Default: 60000
- Fail
Timeout float64 - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - Failover
Criterias []string - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- Latency
Strategy string - What metrics to use for latency. Available values are:
tpot(time-per-output-token) ande2e. possible known values include one of ["e2e", "tpot"]; Default: "tpot" - Max
Fails float64 - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - Read
Timeout float64 - Default: 60000
- Retries float64
- The number of retries to execute upon failure to proxy. Default: 5
- Slots float64
- The number of slots in the load balancer algorithm. Default: 10000
- Write
Timeout float64 - Default: 60000
- connect_
timeout number - Default: 60000
- fail_
timeout number - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover_
criterias list(string) - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- latency_
strategy string - What metrics to use for latency. Available values are:
tpot(time-per-output-token) ande2e. possible known values include one of ["e2e", "tpot"]; Default: "tpot" - max_
fails number - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read_
timeout number - Default: 60000
- retries number
- The number of retries to execute upon failure to proxy. Default: 5
- slots number
- The number of slots in the load balancer algorithm. Default: 10000
- write_
timeout number - Default: 60000
- connect
Timeout Double - Default: 60000
- fail
Timeout Double - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover
Criterias List<String> - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- latency
Strategy String - What metrics to use for latency. Available values are:
tpot(time-per-output-token) ande2e. possible known values include one of ["e2e", "tpot"]; Default: "tpot" - max
Fails Double - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read
Timeout Double - Default: 60000
- retries Double
- The number of retries to execute upon failure to proxy. Default: 5
- slots Double
- The number of slots in the load balancer algorithm. Default: 10000
- write
Timeout Double - Default: 60000
- connect
Timeout number - Default: 60000
- fail
Timeout number - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover
Criterias string[] - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- latency
Strategy string - What metrics to use for latency. Available values are:
tpot(time-per-output-token) ande2e. possible known values include one of ["e2e", "tpot"]; Default: "tpot" - max
Fails number - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read
Timeout number - Default: 60000
- retries number
- The number of retries to execute upon failure to proxy. Default: 5
- slots number
- The number of slots in the load balancer algorithm. Default: 10000
- write
Timeout number - Default: 60000
- connect_
timeout float - Default: 60000
- fail_
timeout float - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover_
criterias Sequence[str] - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- latency_
strategy str - What metrics to use for latency. Available values are:
tpot(time-per-output-token) ande2e. possible known values include one of ["e2e", "tpot"]; Default: "tpot" - max_
fails float - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read_
timeout float - Default: 60000
- retries float
- The number of retries to execute upon failure to proxy. Default: 5
- slots float
- The number of slots in the load balancer algorithm. Default: 10000
- write_
timeout float - Default: 60000
- connect
Timeout Number - Default: 60000
- fail
Timeout Number - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover
Criterias List<String> - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- latency
Strategy String - What metrics to use for latency. Available values are:
tpot(time-per-output-token) ande2e. possible known values include one of ["e2e", "tpot"]; Default: "tpot" - max
Fails Number - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read
Timeout Number - Default: 60000
- retries Number
- The number of retries to execute upon failure to proxy. Default: 5
- slots Number
- The number of slots in the load balancer algorithm. Default: 10000
- write
Timeout Number - Default: 60000
AiGatewayModelModelConfigBalancerLowestUsage, AiGatewayModelModelConfigBalancerLowestUsageArgs
- Connect
Timeout double - Default: 60000
- Fail
Timeout double - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - Failover
Criterias List<string> - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- Max
Fails double - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - Read
Timeout double - Default: 60000
- Retries double
- The number of retries to execute upon failure to proxy. Default: 5
- Slots double
- The number of slots in the load balancer algorithm. Default: 10000
- Tokens
Count stringStrategy - Methodology to use for token usage calculation. possible known values include one of ["completion-tokens", "cost", "llm-accuracy", "prompt-tokens", "total-tokens"]; Default: "total-tokens"
- Write
Timeout double - Default: 60000
- Connect
Timeout float64 - Default: 60000
- Fail
Timeout float64 - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - Failover
Criterias []string - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- Max
Fails float64 - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - Read
Timeout float64 - Default: 60000
- Retries float64
- The number of retries to execute upon failure to proxy. Default: 5
- Slots float64
- The number of slots in the load balancer algorithm. Default: 10000
- Tokens
Count stringStrategy - Methodology to use for token usage calculation. possible known values include one of ["completion-tokens", "cost", "llm-accuracy", "prompt-tokens", "total-tokens"]; Default: "total-tokens"
- Write
Timeout float64 - Default: 60000
- connect_
timeout number - Default: 60000
- fail_
timeout number - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover_
criterias list(string) - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- max_
fails number - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read_
timeout number - Default: 60000
- retries number
- The number of retries to execute upon failure to proxy. Default: 5
- slots number
- The number of slots in the load balancer algorithm. Default: 10000
- tokens_
count_ stringstrategy - Methodology to use for token usage calculation. possible known values include one of ["completion-tokens", "cost", "llm-accuracy", "prompt-tokens", "total-tokens"]; Default: "total-tokens"
- write_
timeout number - Default: 60000
- connect
Timeout Double - Default: 60000
- fail
Timeout Double - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover
Criterias List<String> - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- max
Fails Double - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read
Timeout Double - Default: 60000
- retries Double
- The number of retries to execute upon failure to proxy. Default: 5
- slots Double
- The number of slots in the load balancer algorithm. Default: 10000
- tokens
Count StringStrategy - Methodology to use for token usage calculation. possible known values include one of ["completion-tokens", "cost", "llm-accuracy", "prompt-tokens", "total-tokens"]; Default: "total-tokens"
- write
Timeout Double - Default: 60000
- connect
Timeout number - Default: 60000
- fail
Timeout number - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover
Criterias string[] - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- max
Fails number - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read
Timeout number - Default: 60000
- retries number
- The number of retries to execute upon failure to proxy. Default: 5
- slots number
- The number of slots in the load balancer algorithm. Default: 10000
- tokens
Count stringStrategy - Methodology to use for token usage calculation. possible known values include one of ["completion-tokens", "cost", "llm-accuracy", "prompt-tokens", "total-tokens"]; Default: "total-tokens"
- write
Timeout number - Default: 60000
- connect_
timeout float - Default: 60000
- fail_
timeout float - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover_
criterias Sequence[str] - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- max_
fails float - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read_
timeout float - Default: 60000
- retries float
- The number of retries to execute upon failure to proxy. Default: 5
- slots float
- The number of slots in the load balancer algorithm. Default: 10000
- tokens_
count_ strstrategy - Methodology to use for token usage calculation. possible known values include one of ["completion-tokens", "cost", "llm-accuracy", "prompt-tokens", "total-tokens"]; Default: "total-tokens"
- write_
timeout float - Default: 60000
- connect
Timeout Number - Default: 60000
- fail
Timeout Number - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover
Criterias List<String> - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- max
Fails Number - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read
Timeout Number - Default: 60000
- retries Number
- The number of retries to execute upon failure to proxy. Default: 5
- slots Number
- The number of slots in the load balancer algorithm. Default: 10000
- tokens
Count StringStrategy - Methodology to use for token usage calculation. possible known values include one of ["completion-tokens", "cost", "llm-accuracy", "prompt-tokens", "total-tokens"]; Default: "total-tokens"
- write
Timeout Number - Default: 60000
AiGatewayModelModelConfigBalancerPriority, AiGatewayModelModelConfigBalancerPriorityArgs
- Connect
Timeout double - Default: 60000
- Fail
Timeout double - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - Failover
Criterias List<string> - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- Max
Fails double - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - Read
Timeout double - Default: 60000
- Retries double
- The number of retries to execute upon failure to proxy. Default: 5
- Slots double
- The number of slots in the load balancer algorithm. Default: 10000
- Write
Timeout double - Default: 60000
- Connect
Timeout float64 - Default: 60000
- Fail
Timeout float64 - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - Failover
Criterias []string - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- Max
Fails float64 - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - Read
Timeout float64 - Default: 60000
- Retries float64
- The number of retries to execute upon failure to proxy. Default: 5
- Slots float64
- The number of slots in the load balancer algorithm. Default: 10000
- Write
Timeout float64 - Default: 60000
- connect_
timeout number - Default: 60000
- fail_
timeout number - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover_
criterias list(string) - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- max_
fails number - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read_
timeout number - Default: 60000
- retries number
- The number of retries to execute upon failure to proxy. Default: 5
- slots number
- The number of slots in the load balancer algorithm. Default: 10000
- write_
timeout number - Default: 60000
- connect
Timeout Double - Default: 60000
- fail
Timeout Double - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover
Criterias List<String> - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- max
Fails Double - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read
Timeout Double - Default: 60000
- retries Double
- The number of retries to execute upon failure to proxy. Default: 5
- slots Double
- The number of slots in the load balancer algorithm. Default: 10000
- write
Timeout Double - Default: 60000
- connect
Timeout number - Default: 60000
- fail
Timeout number - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover
Criterias string[] - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- max
Fails number - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read
Timeout number - Default: 60000
- retries number
- The number of retries to execute upon failure to proxy. Default: 5
- slots number
- The number of slots in the load balancer algorithm. Default: 10000
- write
Timeout number - Default: 60000
- connect_
timeout float - Default: 60000
- fail_
timeout float - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover_
criterias Sequence[str] - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- max_
fails float - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read_
timeout float - Default: 60000
- retries float
- The number of retries to execute upon failure to proxy. Default: 5
- slots float
- The number of slots in the load balancer algorithm. Default: 10000
- write_
timeout float - Default: 60000
- connect
Timeout Number - Default: 60000
- fail
Timeout Number - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover
Criterias List<String> - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- max
Fails Number - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read
Timeout Number - Default: 60000
- retries Number
- The number of retries to execute upon failure to proxy. Default: 5
- slots Number
- The number of slots in the load balancer algorithm. Default: 10000
- write
Timeout Number - Default: 60000
AiGatewayModelModelConfigBalancerRoundRobin, AiGatewayModelModelConfigBalancerRoundRobinArgs
- Connect
Timeout double - Default: 60000
- Fail
Timeout double - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - Failover
Criterias List<string> - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- Max
Fails double - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - Read
Timeout double - Default: 60000
- Retries double
- The number of retries to execute upon failure to proxy. Default: 5
- Slots double
- The number of slots in the load balancer algorithm. Default: 10000
- Write
Timeout double - Default: 60000
- Connect
Timeout float64 - Default: 60000
- Fail
Timeout float64 - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - Failover
Criterias []string - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- Max
Fails float64 - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - Read
Timeout float64 - Default: 60000
- Retries float64
- The number of retries to execute upon failure to proxy. Default: 5
- Slots float64
- The number of slots in the load balancer algorithm. Default: 10000
- Write
Timeout float64 - Default: 60000
- connect_
timeout number - Default: 60000
- fail_
timeout number - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover_
criterias list(string) - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- max_
fails number - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read_
timeout number - Default: 60000
- retries number
- The number of retries to execute upon failure to proxy. Default: 5
- slots number
- The number of slots in the load balancer algorithm. Default: 10000
- write_
timeout number - Default: 60000
- connect
Timeout Double - Default: 60000
- fail
Timeout Double - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover
Criterias List<String> - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- max
Fails Double - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read
Timeout Double - Default: 60000
- retries Double
- The number of retries to execute upon failure to proxy. Default: 5
- slots Double
- The number of slots in the load balancer algorithm. Default: 10000
- write
Timeout Double - Default: 60000
- connect
Timeout number - Default: 60000
- fail
Timeout number - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover
Criterias string[] - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- max
Fails number - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read
Timeout number - Default: 60000
- retries number
- The number of retries to execute upon failure to proxy. Default: 5
- slots number
- The number of slots in the load balancer algorithm. Default: 10000
- write
Timeout number - Default: 60000
- connect_
timeout float - Default: 60000
- fail_
timeout float - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover_
criterias Sequence[str] - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- max_
fails float - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read_
timeout float - Default: 60000
- retries float
- The number of retries to execute upon failure to proxy. Default: 5
- slots float
- The number of slots in the load balancer algorithm. Default: 10000
- write_
timeout float - Default: 60000
- connect
Timeout Number - Default: 60000
- fail
Timeout Number - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover
Criterias List<String> - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- max
Fails Number - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read
Timeout Number - Default: 60000
- retries Number
- The number of retries to execute upon failure to proxy. Default: 5
- slots Number
- The number of slots in the load balancer algorithm. Default: 10000
- write
Timeout Number - Default: 60000
AiGatewayModelModelConfigBalancerSemantic, AiGatewayModelModelConfigBalancerSemanticArgs
- Connect
Timeout double - Default: 60000
- Embeddings
Ai
Gateway Model Model Config Balancer Semantic Embeddings - Embeddings model configuration for this model. Not Null
- Fail
Timeout double - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - Failover
Criterias List<string> - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- Max
Fails double - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - Read
Timeout double - Default: 60000
- Retries double
- The number of retries to execute upon failure to proxy. Default: 5
- Slots double
- The number of slots in the load balancer algorithm. Default: 10000
- Vectordb
Ai
Gateway Model Model Config Balancer Semantic Vectordb - Configuration for the vector database used by the model. Not Null
- Write
Timeout double - Default: 60000
- Connect
Timeout float64 - Default: 60000
- Embeddings
Ai
Gateway Model Model Config Balancer Semantic Embeddings - Embeddings model configuration for this model. Not Null
- Fail
Timeout float64 - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - Failover
Criterias []string - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- Max
Fails float64 - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - Read
Timeout float64 - Default: 60000
- Retries float64
- The number of retries to execute upon failure to proxy. Default: 5
- Slots float64
- The number of slots in the load balancer algorithm. Default: 10000
- Vectordb
Ai
Gateway Model Model Config Balancer Semantic Vectordb - Configuration for the vector database used by the model. Not Null
- Write
Timeout float64 - Default: 60000
- connect_
timeout number - Default: 60000
- embeddings object
- Embeddings model configuration for this model. Not Null
- fail_
timeout number - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover_
criterias list(string) - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- max_
fails number - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read_
timeout number - Default: 60000
- retries number
- The number of retries to execute upon failure to proxy. Default: 5
- slots number
- The number of slots in the load balancer algorithm. Default: 10000
- vectordb object
- Configuration for the vector database used by the model. Not Null
- write_
timeout number - Default: 60000
- connect
Timeout Double - Default: 60000
- embeddings
Ai
Gateway Model Model Config Balancer Semantic Embeddings - Embeddings model configuration for this model. Not Null
- fail
Timeout Double - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover
Criterias List<String> - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- max
Fails Double - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read
Timeout Double - Default: 60000
- retries Double
- The number of retries to execute upon failure to proxy. Default: 5
- slots Double
- The number of slots in the load balancer algorithm. Default: 10000
- vectordb
Ai
Gateway Model Model Config Balancer Semantic Vectordb - Configuration for the vector database used by the model. Not Null
- write
Timeout Double - Default: 60000
- connect
Timeout number - Default: 60000
- embeddings
Ai
Gateway Model Model Config Balancer Semantic Embeddings - Embeddings model configuration for this model. Not Null
- fail
Timeout number - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover
Criterias string[] - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- max
Fails number - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read
Timeout number - Default: 60000
- retries number
- The number of retries to execute upon failure to proxy. Default: 5
- slots number
- The number of slots in the load balancer algorithm. Default: 10000
- vectordb
Ai
Gateway Model Model Config Balancer Semantic Vectordb - Configuration for the vector database used by the model. Not Null
- write
Timeout number - Default: 60000
- connect_
timeout float - Default: 60000
- embeddings
Ai
Gateway Model Model Config Balancer Semantic Embeddings - Embeddings model configuration for this model. Not Null
- fail_
timeout float - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover_
criterias Sequence[str] - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- max_
fails float - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read_
timeout float - Default: 60000
- retries float
- The number of retries to execute upon failure to proxy. Default: 5
- slots float
- The number of slots in the load balancer algorithm. Default: 10000
- vectordb
Ai
Gateway Model Model Config Balancer Semantic Vectordb - Configuration for the vector database used by the model. Not Null
- write_
timeout float - Default: 60000
- connect
Timeout Number - Default: 60000
- embeddings Property Map
- Embeddings model configuration for this model. Not Null
- fail
Timeout Number - The period of time (in milliseconds) the target will be considered unavailable after the number of unsuccessful attempts reaches
max_fails. Default: 10000 - failover
Criterias List<String> - Specifies in which cases an upstream response should be failover to the next target. Each option in the array is equivalent to the function of https://nginx.org/en/docs/http/ngxhttpproxymodule.html#proxynext_upstream. Default: ["error","timeout"]
- max
Fails Number - Number of unsuccessful attempts to communicate with a target that should occur in the duration defined by
fail_timeoutbefore the target is considered unavailable. The zero value disables the circuit breaker. What is considered an unsuccessful attempt is defined byfailover_criteria. Note the cases oferror,timeoutandinvalid_headerare always considered unsuccessful attempts, while the cases ofhttp_403andhttp_404are never considered unsuccessful attempts. Default: 0 - read
Timeout Number - Default: 60000
- retries Number
- The number of retries to execute upon failure to proxy. Default: 5
- slots Number
- The number of slots in the load balancer algorithm. Default: 10000
- vectordb Property Map
- Configuration for the vector database used by the model. Not Null
- write
Timeout Number - Default: 60000
AiGatewayModelModelConfigBalancerSemanticEmbeddings, AiGatewayModelModelConfigBalancerSemanticEmbeddingsArgs
- Allow
Auth boolOverride - When enabled, request-level auth parameters (such as API keys or bearer tokens) will override the static values defined for the provider. Default: false
- Config
Ai
Gateway Model Model Config Balancer Semantic Embeddings Config - Configuration for an embeddings model. Not Null
- Name string
- The name of the embeddings model. Not Null
- Provider string
- Reference to a model provider instance by name. Not Null
- Allow
Auth boolOverride - When enabled, request-level auth parameters (such as API keys or bearer tokens) will override the static values defined for the provider. Default: false
- Config
Ai
Gateway Model Model Config Balancer Semantic Embeddings Config - Configuration for an embeddings model. Not Null
- Name string
- The name of the embeddings model. Not Null
- Provider string
- Reference to a model provider instance by name. Not Null
- allow_
auth_ booloverride - When enabled, request-level auth parameters (such as API keys or bearer tokens) will override the static values defined for the provider. Default: false
- config object
- Configuration for an embeddings model. Not Null
- name string
- The name of the embeddings model. Not Null
- provider string
- Reference to a model provider instance by name. Not Null
- allow
Auth BooleanOverride - When enabled, request-level auth parameters (such as API keys or bearer tokens) will override the static values defined for the provider. Default: false
- config
Ai
Gateway Model Model Config Balancer Semantic Embeddings Config - Configuration for an embeddings model. Not Null
- name String
- The name of the embeddings model. Not Null
- provider String
- Reference to a model provider instance by name. Not Null
- allow
Auth booleanOverride - When enabled, request-level auth parameters (such as API keys or bearer tokens) will override the static values defined for the provider. Default: false
- config
Ai
Gateway Model Model Config Balancer Semantic Embeddings Config - Configuration for an embeddings model. Not Null
- name string
- The name of the embeddings model. Not Null
- provider string
- Reference to a model provider instance by name. Not Null
- allow_
auth_ booloverride - When enabled, request-level auth parameters (such as API keys or bearer tokens) will override the static values defined for the provider. Default: false
- config
Ai
Gateway Model Model Config Balancer Semantic Embeddings Config - Configuration for an embeddings model. Not Null
- name str
- The name of the embeddings model. Not Null
- provider str
- Reference to a model provider instance by name. Not Null
- allow
Auth BooleanOverride - When enabled, request-level auth parameters (such as API keys or bearer tokens) will override the static values defined for the provider. Default: false
- config Property Map
- Configuration for an embeddings model. Not Null
- name String
- The name of the embeddings model. Not Null
- provider String
- Reference to a model provider instance by name. Not Null
AiGatewayModelModelConfigBalancerSemanticEmbeddingsConfig, AiGatewayModelModelConfigBalancerSemanticEmbeddingsConfigArgs
- Azure
Ai
Gateway Model Model Config Balancer Semantic Embeddings Config Azure - Azure OpenAI-specific configuration for an embeddings model. Azure AI Foundry embeddings are not supported.
- Bedrock
Ai
Gateway Model Model Config Balancer Semantic Embeddings Config Bedrock - AWS Bedrock-specific configuration for a model.
- Gemini
Ai
Gateway Model Model Config Balancer Semantic Embeddings Config Gemini - Google Gemini-specific configuration for a model.
- Huggingface
Ai
Gateway Model Model Config Balancer Semantic Embeddings Config Huggingface - Hugging Face-specific configuration for a model.
- Mistral
Ai
Gateway Model Model Config Balancer Semantic Embeddings Config Mistral - Ollama
Ai
Gateway Model Model Config Balancer Semantic Embeddings Config Ollama - Ollama-specific configuration for a model.
- Openai
Ai
Gateway Model Model Config Balancer Semantic Embeddings Config Openai - Openai-specific configuration for a model.
- Azure
Ai
Gateway Model Model Config Balancer Semantic Embeddings Config Azure - Azure OpenAI-specific configuration for an embeddings model. Azure AI Foundry embeddings are not supported.
- Bedrock
Ai
Gateway Model Model Config Balancer Semantic Embeddings Config Bedrock - AWS Bedrock-specific configuration for a model.
- Gemini
Ai
Gateway Model Model Config Balancer Semantic Embeddings Config Gemini - Google Gemini-specific configuration for a model.
- Huggingface
Ai
Gateway Model Model Config Balancer Semantic Embeddings Config Huggingface - Hugging Face-specific configuration for a model.
- Mistral
Ai
Gateway Model Model Config Balancer Semantic Embeddings Config Mistral - Ollama
Ai
Gateway Model Model Config Balancer Semantic Embeddings Config Ollama - Ollama-specific configuration for a model.
- Openai
Ai
Gateway Model Model Config Balancer Semantic Embeddings Config Openai - Openai-specific configuration for a model.
- azure object
- Azure OpenAI-specific configuration for an embeddings model. Azure AI Foundry embeddings are not supported.
- bedrock object
- AWS Bedrock-specific configuration for a model.
- gemini object
- Google Gemini-specific configuration for a model.
- huggingface object
- Hugging Face-specific configuration for a model.
- mistral object
- ollama object
- Ollama-specific configuration for a model.
- openai object
- Openai-specific configuration for a model.
- azure
Ai
Gateway Model Model Config Balancer Semantic Embeddings Config Azure - Azure OpenAI-specific configuration for an embeddings model. Azure AI Foundry embeddings are not supported.
- bedrock
Ai
Gateway Model Model Config Balancer Semantic Embeddings Config Bedrock - AWS Bedrock-specific configuration for a model.
- gemini
Ai
Gateway Model Model Config Balancer Semantic Embeddings Config Gemini - Google Gemini-specific configuration for a model.
- huggingface
Ai
Gateway Model Model Config Balancer Semantic Embeddings Config Huggingface - Hugging Face-specific configuration for a model.
- mistral
Ai
Gateway Model Model Config Balancer Semantic Embeddings Config Mistral - ollama
Ai
Gateway Model Model Config Balancer Semantic Embeddings Config Ollama - Ollama-specific configuration for a model.
- openai
Ai
Gateway Model Model Config Balancer Semantic Embeddings Config Openai - Openai-specific configuration for a model.
- azure
Ai
Gateway Model Model Config Balancer Semantic Embeddings Config Azure - Azure OpenAI-specific configuration for an embeddings model. Azure AI Foundry embeddings are not supported.
- bedrock
Ai
Gateway Model Model Config Balancer Semantic Embeddings Config Bedrock - AWS Bedrock-specific configuration for a model.
- gemini
Ai
Gateway Model Model Config Balancer Semantic Embeddings Config Gemini - Google Gemini-specific configuration for a model.
- huggingface
Ai
Gateway Model Model Config Balancer Semantic Embeddings Config Huggingface - Hugging Face-specific configuration for a model.
- mistral
Ai
Gateway Model Model Config Balancer Semantic Embeddings Config Mistral - ollama
Ai
Gateway Model Model Config Balancer Semantic Embeddings Config Ollama - Ollama-specific configuration for a model.
- openai
Ai
Gateway Model Model Config Balancer Semantic Embeddings Config Openai - Openai-specific configuration for a model.
- azure
Ai
Gateway Model Model Config Balancer Semantic Embeddings Config Azure - Azure OpenAI-specific configuration for an embeddings model. Azure AI Foundry embeddings are not supported.
- bedrock
Ai
Gateway Model Model Config Balancer Semantic Embeddings Config Bedrock - AWS Bedrock-specific configuration for a model.
- gemini
Ai
Gateway Model Model Config Balancer Semantic Embeddings Config Gemini - Google Gemini-specific configuration for a model.
- huggingface
Ai
Gateway Model Model Config Balancer Semantic Embeddings Config Huggingface - Hugging Face-specific configuration for a model.
- mistral
Ai
Gateway Model Model Config Balancer Semantic Embeddings Config Mistral - ollama
Ai
Gateway Model Model Config Balancer Semantic Embeddings Config Ollama - Ollama-specific configuration for a model.
- openai
Ai
Gateway Model Model Config Balancer Semantic Embeddings Config Openai - Openai-specific configuration for a model.
- azure Property Map
- Azure OpenAI-specific configuration for an embeddings model. Azure AI Foundry embeddings are not supported.
- bedrock Property Map
- AWS Bedrock-specific configuration for a model.
- gemini Property Map
- Google Gemini-specific configuration for a model.
- huggingface Property Map
- Hugging Face-specific configuration for a model.
- mistral Property Map
- ollama Property Map
- Ollama-specific configuration for a model.
- openai Property Map
- Openai-specific configuration for a model.
AiGatewayModelModelConfigBalancerSemanticEmbeddingsConfigAzure, AiGatewayModelModelConfigBalancerSemanticEmbeddingsConfigAzureArgs
- Api
Version string - The Azure OpenAI API version to use. Default: "2023-05-15"
- Deployment
Id string - The Azure OpenAI deployment ID for the embeddings model. Not Null
- Type string
- Not Null; must be "azure"
- Upstream
Url string - The URL of the embeddings model.
- Api
Version string - The Azure OpenAI API version to use. Default: "2023-05-15"
- Deployment
Id string - The Azure OpenAI deployment ID for the embeddings model. Not Null
- Type string
- Not Null; must be "azure"
- Upstream
Url string - The URL of the embeddings model.
- api_
version string - The Azure OpenAI API version to use. Default: "2023-05-15"
- deployment_
id string - The Azure OpenAI deployment ID for the embeddings model. Not Null
- type string
- Not Null; must be "azure"
- upstream_
url string - The URL of the embeddings model.
- api
Version String - The Azure OpenAI API version to use. Default: "2023-05-15"
- deployment
Id String - The Azure OpenAI deployment ID for the embeddings model. Not Null
- type String
- Not Null; must be "azure"
- upstream
Url String - The URL of the embeddings model.
- api
Version string - The Azure OpenAI API version to use. Default: "2023-05-15"
- deployment
Id string - The Azure OpenAI deployment ID for the embeddings model. Not Null
- type string
- Not Null; must be "azure"
- upstream
Url string - The URL of the embeddings model.
- api_
version str - The Azure OpenAI API version to use. Default: "2023-05-15"
- deployment_
id str - The Azure OpenAI deployment ID for the embeddings model. Not Null
- type str
- Not Null; must be "azure"
- upstream_
url str - The URL of the embeddings model.
- api
Version String - The Azure OpenAI API version to use. Default: "2023-05-15"
- deployment
Id String - The Azure OpenAI deployment ID for the embeddings model. Not Null
- type String
- Not Null; must be "azure"
- upstream
Url String - The URL of the embeddings model.
AiGatewayModelModelConfigBalancerSemanticEmbeddingsConfigBedrock, AiGatewayModelModelConfigBalancerSemanticEmbeddingsConfigBedrockArgs
- Batch
Bucket stringPrefix - S3 bucket prefix for batch inference jobs.
- Embeddings
Normalize bool - Whether to normalize embedding vectors in the response. Default: false
- Performance
Config stringLatency - Latency performance configuration for the model invocation.
- Region string
- The AWS region for the model. Setting this option overrides the AWS_REGION environment variable.
- Upstream
Url string - The URL of the embeddings model.
- Video
Output stringS3Uri - S3 URI for storing video generation outputs.
- Batch
Bucket stringPrefix - S3 bucket prefix for batch inference jobs.
- Embeddings
Normalize bool - Whether to normalize embedding vectors in the response. Default: false
- Performance
Config stringLatency - Latency performance configuration for the model invocation.
- Region string
- The AWS region for the model. Setting this option overrides the AWS_REGION environment variable.
- Upstream
Url string - The URL of the embeddings model.
- Video
Output stringS3Uri - S3 URI for storing video generation outputs.
- batch_
bucket_ stringprefix - S3 bucket prefix for batch inference jobs.
- embeddings_
normalize bool - Whether to normalize embedding vectors in the response. Default: false
- performance_
config_ stringlatency - Latency performance configuration for the model invocation.
- region string
- The AWS region for the model. Setting this option overrides the AWS_REGION environment variable.
- upstream_
url string - The URL of the embeddings model.
- video_
output_ strings3_ uri - S3 URI for storing video generation outputs.
- batch
Bucket StringPrefix - S3 bucket prefix for batch inference jobs.
- embeddings
Normalize Boolean - Whether to normalize embedding vectors in the response. Default: false
- performance
Config StringLatency - Latency performance configuration for the model invocation.
- region String
- The AWS region for the model. Setting this option overrides the AWS_REGION environment variable.
- upstream
Url String - The URL of the embeddings model.
- video
Output StringS3Uri - S3 URI for storing video generation outputs.
- batch
Bucket stringPrefix - S3 bucket prefix for batch inference jobs.
- embeddings
Normalize boolean - Whether to normalize embedding vectors in the response. Default: false
- performance
Config stringLatency - Latency performance configuration for the model invocation.
- region string
- The AWS region for the model. Setting this option overrides the AWS_REGION environment variable.
- upstream
Url string - The URL of the embeddings model.
- video
Output stringS3Uri - S3 URI for storing video generation outputs.
- batch_
bucket_ strprefix - S3 bucket prefix for batch inference jobs.
- embeddings_
normalize bool - Whether to normalize embedding vectors in the response. Default: false
- performance_
config_ strlatency - Latency performance configuration for the model invocation.
- region str
- The AWS region for the model. Setting this option overrides the AWS_REGION environment variable.
- upstream_
url str - The URL of the embeddings model.
- video_
output_ strs3_ uri - S3 URI for storing video generation outputs.
- batch
Bucket StringPrefix - S3 bucket prefix for batch inference jobs.
- embeddings
Normalize Boolean - Whether to normalize embedding vectors in the response. Default: false
- performance
Config StringLatency - Latency performance configuration for the model invocation.
- region String
- The AWS region for the model. Setting this option overrides the AWS_REGION environment variable.
- upstream
Url String - The URL of the embeddings model.
- video
Output StringS3Uri - S3 URI for storing video generation outputs.
AiGatewayModelModelConfigBalancerSemanticEmbeddingsConfigGemini, AiGatewayModelModelConfigBalancerSemanticEmbeddingsConfigGeminiArgs
- Gcp
Environment AiGateway Model Model Config Balancer Semantic Embeddings Config Gemini Gcp Environment - Configuration for a model hosted on Google Cloud Project.
- Upstream
Url string - The URL of the embeddings model.
- Gcp
Environment AiGateway Model Model Config Balancer Semantic Embeddings Config Gemini Gcp Environment - Configuration for a model hosted on Google Cloud Project.
- Upstream
Url string - The URL of the embeddings model.
- gcp_
environment object - Configuration for a model hosted on Google Cloud Project.
- upstream_
url string - The URL of the embeddings model.
- gcp
Environment AiGateway Model Model Config Balancer Semantic Embeddings Config Gemini Gcp Environment - Configuration for a model hosted on Google Cloud Project.
- upstream
Url String - The URL of the embeddings model.
- gcp
Environment AiGateway Model Model Config Balancer Semantic Embeddings Config Gemini Gcp Environment - Configuration for a model hosted on Google Cloud Project.
- upstream
Url string - The URL of the embeddings model.
- gcp_
environment AiGateway Model Model Config Balancer Semantic Embeddings Config Gemini Gcp Environment - Configuration for a model hosted on Google Cloud Project.
- upstream_
url str - The URL of the embeddings model.
- gcp
Environment Property Map - Configuration for a model hosted on Google Cloud Project.
- upstream
Url String - The URL of the embeddings model.
AiGatewayModelModelConfigBalancerSemanticEmbeddingsConfigGeminiGcpEnvironment, AiGatewayModelModelConfigBalancerSemanticEmbeddingsConfigGeminiGcpEnvironmentArgs
- Api
Endpoint string - The custom API endpoint for the Gemini model. Not Null
- Location
Id string - The Google Cloud location ID for the model endpoint. Not Null
- Project
Id string - The Google Cloud project ID for the model endpoint. Not Null
- Api
Endpoint string - The custom API endpoint for the Gemini model. Not Null
- Location
Id string - The Google Cloud location ID for the model endpoint. Not Null
- Project
Id string - The Google Cloud project ID for the model endpoint. Not Null
- api_
endpoint string - The custom API endpoint for the Gemini model. Not Null
- location_
id string - The Google Cloud location ID for the model endpoint. Not Null
- project_
id string - The Google Cloud project ID for the model endpoint. Not Null
- api
Endpoint String - The custom API endpoint for the Gemini model. Not Null
- location
Id String - The Google Cloud location ID for the model endpoint. Not Null
- project
Id String - The Google Cloud project ID for the model endpoint. Not Null
- api
Endpoint string - The custom API endpoint for the Gemini model. Not Null
- location
Id string - The Google Cloud location ID for the model endpoint. Not Null
- project
Id string - The Google Cloud project ID for the model endpoint. Not Null
- api_
endpoint str - The custom API endpoint for the Gemini model. Not Null
- location_
id str - The Google Cloud location ID for the model endpoint. Not Null
- project_
id str - The Google Cloud project ID for the model endpoint. Not Null
- api
Endpoint String - The custom API endpoint for the Gemini model. Not Null
- location
Id String - The Google Cloud location ID for the model endpoint. Not Null
- project
Id String - The Google Cloud project ID for the model endpoint. Not Null
AiGatewayModelModelConfigBalancerSemanticEmbeddingsConfigHuggingface, AiGatewayModelModelConfigBalancerSemanticEmbeddingsConfigHuggingfaceArgs
- Upstream
Url string - The URL of the embeddings model.
- Use
Cache bool - Whether to use the Hugging Face inference cache. Default: false
- Wait
For boolModel - Whether to wait for the model to load if it is not ready. Default: false
- Upstream
Url string - The URL of the embeddings model.
- Use
Cache bool - Whether to use the Hugging Face inference cache. Default: false
- Wait
For boolModel - Whether to wait for the model to load if it is not ready. Default: false
- upstream_
url string - The URL of the embeddings model.
- use_
cache bool - Whether to use the Hugging Face inference cache. Default: false
- wait_
for_ boolmodel - Whether to wait for the model to load if it is not ready. Default: false
- upstream
Url String - The URL of the embeddings model.
- use
Cache Boolean - Whether to use the Hugging Face inference cache. Default: false
- wait
For BooleanModel - Whether to wait for the model to load if it is not ready. Default: false
- upstream
Url string - The URL of the embeddings model.
- use
Cache boolean - Whether to use the Hugging Face inference cache. Default: false
- wait
For booleanModel - Whether to wait for the model to load if it is not ready. Default: false
- upstream_
url str - The URL of the embeddings model.
- use_
cache bool - Whether to use the Hugging Face inference cache. Default: false
- wait_
for_ boolmodel - Whether to wait for the model to load if it is not ready. Default: false
- upstream
Url String - The URL of the embeddings model.
- use
Cache Boolean - Whether to use the Hugging Face inference cache. Default: false
- wait
For BooleanModel - Whether to wait for the model to load if it is not ready. Default: false
AiGatewayModelModelConfigBalancerSemanticEmbeddingsConfigMistral, AiGatewayModelModelConfigBalancerSemanticEmbeddingsConfigMistralArgs
- Type string
- Not Null; must be "mistral"
- Upstream
Url string - The URL of the embeddings model.
- Type string
- Not Null; must be "mistral"
- Upstream
Url string - The URL of the embeddings model.
- type string
- Not Null; must be "mistral"
- upstream_
url string - The URL of the embeddings model.
- type String
- Not Null; must be "mistral"
- upstream
Url String - The URL of the embeddings model.
- type string
- Not Null; must be "mistral"
- upstream
Url string - The URL of the embeddings model.
- type str
- Not Null; must be "mistral"
- upstream_
url str - The URL of the embeddings model.
- type String
- Not Null; must be "mistral"
- upstream
Url String - The URL of the embeddings model.
AiGatewayModelModelConfigBalancerSemanticEmbeddingsConfigOllama, AiGatewayModelModelConfigBalancerSemanticEmbeddingsConfigOllamaArgs
- Upstream
Url string - The URL of the embeddings model.
- Upstream
Url string - The URL of the embeddings model.
- upstream_
url string - The URL of the embeddings model.
- upstream
Url String - The URL of the embeddings model.
- upstream
Url string - The URL of the embeddings model.
- upstream_
url str - The URL of the embeddings model.
- upstream
Url String - The URL of the embeddings model.
AiGatewayModelModelConfigBalancerSemanticEmbeddingsConfigOpenai, AiGatewayModelModelConfigBalancerSemanticEmbeddingsConfigOpenaiArgs
- Upstream
Url string - The URL of the embeddings model.
- Upstream
Url string - The URL of the embeddings model.
- upstream_
url string - The URL of the embeddings model.
- upstream
Url String - The URL of the embeddings model.
- upstream
Url string - The URL of the embeddings model.
- upstream_
url str - The URL of the embeddings model.
- upstream
Url String - The URL of the embeddings model.
AiGatewayModelModelConfigBalancerSemanticVectordb, AiGatewayModelModelConfigBalancerSemanticVectordbArgs
- Pgvector
Ai
Gateway Model Model Config Balancer Semantic Vectordb Pgvector - Redis
Ai
Gateway Model Model Config Balancer Semantic Vectordb Redis - Config for connecting to a Cloud Provider's Redis instance.
- Pgvector
Ai
Gateway Model Model Config Balancer Semantic Vectordb Pgvector - Redis
Ai
Gateway Model Model Config Balancer Semantic Vectordb Redis - Config for connecting to a Cloud Provider's Redis instance.
- pgvector
Ai
Gateway Model Model Config Balancer Semantic Vectordb Pgvector - redis
Ai
Gateway Model Model Config Balancer Semantic Vectordb Redis - Config for connecting to a Cloud Provider's Redis instance.
- pgvector
Ai
Gateway Model Model Config Balancer Semantic Vectordb Pgvector - redis
Ai
Gateway Model Model Config Balancer Semantic Vectordb Redis - Config for connecting to a Cloud Provider's Redis instance.
- pgvector
Ai
Gateway Model Model Config Balancer Semantic Vectordb Pgvector - redis
Ai
Gateway Model Model Config Balancer Semantic Vectordb Redis - Config for connecting to a Cloud Provider's Redis instance.
- pgvector Property Map
- redis Property Map
- Config for connecting to a Cloud Provider's Redis instance.
AiGatewayModelModelConfigBalancerSemanticVectordbPgvector, AiGatewayModelModelConfigBalancerSemanticVectordbPgvectorArgs
- Database string
- the database of the pgvector database. Default: "kong-pgvector"
- Dimensions double
- the desired dimensionality for the vectors. Not Null
- Distance
Metric string - the distance metric to use for vector searches. possible known values include one of ["cosine", "euclidean"]; Not Null
- Host string
- the host of the pgvector database. Default: "127.0.0.1"
- Password string
- the password of the pgvector database This field is referenceable.
- Port double
- the port of the pgvector database. Default: 5432
- Ssl
Ai
Gateway Model Model Config Balancer Semantic Vectordb Pgvector Ssl - Threshold double
- the default similarity threshold for accepting semantic search results (float). Higher threshold means more results are considered similar.
- Timeout double
- the timeout of the pgvector database. Default: 5000
- User string
- the user of the pgvector database This field is referenceable. Default: "postgres"
- Database string
- the database of the pgvector database. Default: "kong-pgvector"
- Dimensions float64
- the desired dimensionality for the vectors. Not Null
- Distance
Metric string - the distance metric to use for vector searches. possible known values include one of ["cosine", "euclidean"]; Not Null
- Host string
- the host of the pgvector database. Default: "127.0.0.1"
- Password string
- the password of the pgvector database This field is referenceable.
- Port float64
- the port of the pgvector database. Default: 5432
- Ssl
Ai
Gateway Model Model Config Balancer Semantic Vectordb Pgvector Ssl - Threshold float64
- the default similarity threshold for accepting semantic search results (float). Higher threshold means more results are considered similar.
- Timeout float64
- the timeout of the pgvector database. Default: 5000
- User string
- the user of the pgvector database This field is referenceable. Default: "postgres"
- database string
- the database of the pgvector database. Default: "kong-pgvector"
- dimensions number
- the desired dimensionality for the vectors. Not Null
- distance_
metric string - the distance metric to use for vector searches. possible known values include one of ["cosine", "euclidean"]; Not Null
- host string
- the host of the pgvector database. Default: "127.0.0.1"
- password string
- the password of the pgvector database This field is referenceable.
- port number
- the port of the pgvector database. Default: 5432
- ssl object
- threshold number
- the default similarity threshold for accepting semantic search results (float). Higher threshold means more results are considered similar.
- timeout number
- the timeout of the pgvector database. Default: 5000
- user string
- the user of the pgvector database This field is referenceable. Default: "postgres"
- database String
- the database of the pgvector database. Default: "kong-pgvector"
- dimensions Double
- the desired dimensionality for the vectors. Not Null
- distance
Metric String - the distance metric to use for vector searches. possible known values include one of ["cosine", "euclidean"]; Not Null
- host String
- the host of the pgvector database. Default: "127.0.0.1"
- password String
- the password of the pgvector database This field is referenceable.
- port Double
- the port of the pgvector database. Default: 5432
- ssl
Ai
Gateway Model Model Config Balancer Semantic Vectordb Pgvector Ssl - threshold Double
- the default similarity threshold for accepting semantic search results (float). Higher threshold means more results are considered similar.
- timeout Double
- the timeout of the pgvector database. Default: 5000
- user String
- the user of the pgvector database This field is referenceable. Default: "postgres"
- database string
- the database of the pgvector database. Default: "kong-pgvector"
- dimensions number
- the desired dimensionality for the vectors. Not Null
- distance
Metric string - the distance metric to use for vector searches. possible known values include one of ["cosine", "euclidean"]; Not Null
- host string
- the host of the pgvector database. Default: "127.0.0.1"
- password string
- the password of the pgvector database This field is referenceable.
- port number
- the port of the pgvector database. Default: 5432
- ssl
Ai
Gateway Model Model Config Balancer Semantic Vectordb Pgvector Ssl - threshold number
- the default similarity threshold for accepting semantic search results (float). Higher threshold means more results are considered similar.
- timeout number
- the timeout of the pgvector database. Default: 5000
- user string
- the user of the pgvector database This field is referenceable. Default: "postgres"
- database str
- the database of the pgvector database. Default: "kong-pgvector"
- dimensions float
- the desired dimensionality for the vectors. Not Null
- distance_
metric str - the distance metric to use for vector searches. possible known values include one of ["cosine", "euclidean"]; Not Null
- host str
- the host of the pgvector database. Default: "127.0.0.1"
- password str
- the password of the pgvector database This field is referenceable.
- port float
- the port of the pgvector database. Default: 5432
- ssl
Ai
Gateway Model Model Config Balancer Semantic Vectordb Pgvector Ssl - threshold float
- the default similarity threshold for accepting semantic search results (float). Higher threshold means more results are considered similar.
- timeout float
- the timeout of the pgvector database. Default: 5000
- user str
- the user of the pgvector database This field is referenceable. Default: "postgres"
- database String
- the database of the pgvector database. Default: "kong-pgvector"
- dimensions Number
- the desired dimensionality for the vectors. Not Null
- distance
Metric String - the distance metric to use for vector searches. possible known values include one of ["cosine", "euclidean"]; Not Null
- host String
- the host of the pgvector database. Default: "127.0.0.1"
- password String
- the password of the pgvector database This field is referenceable.
- port Number
- the port of the pgvector database. Default: 5432
- ssl Property Map
- threshold Number
- the default similarity threshold for accepting semantic search results (float). Higher threshold means more results are considered similar.
- timeout Number
- the timeout of the pgvector database. Default: 5000
- user String
- the user of the pgvector database This field is referenceable. Default: "postgres"
AiGatewayModelModelConfigBalancerSemanticVectordbPgvectorSsl, AiGatewayModelModelConfigBalancerSemanticVectordbPgvectorSslArgs
- Cert string
- the path of ssl cert to use for the pgvector database
- Cert
Key string - the path of ssl cert key to use for the pgvector database
- Enabled bool
- whether to use ssl for the pgvector database. Default: true
- Required bool
- whether ssl is required for the pgvector database. Default: true
- Verify bool
- whether to verify ssl for the pgvector database. Default: true
- Version string
- the ssl version to use for the pgvector database. possible known values include one of ["any", "tlsv12", "tlsv13"]; Default: "tlsv1_2"
- Cert string
- the path of ssl cert to use for the pgvector database
- Cert
Key string - the path of ssl cert key to use for the pgvector database
- Enabled bool
- whether to use ssl for the pgvector database. Default: true
- Required bool
- whether ssl is required for the pgvector database. Default: true
- Verify bool
- whether to verify ssl for the pgvector database. Default: true
- Version string
- the ssl version to use for the pgvector database. possible known values include one of ["any", "tlsv12", "tlsv13"]; Default: "tlsv1_2"
- cert string
- the path of ssl cert to use for the pgvector database
- cert_
key string - the path of ssl cert key to use for the pgvector database
- enabled bool
- whether to use ssl for the pgvector database. Default: true
- required bool
- whether ssl is required for the pgvector database. Default: true
- verify bool
- whether to verify ssl for the pgvector database. Default: true
- version string
- the ssl version to use for the pgvector database. possible known values include one of ["any", "tlsv12", "tlsv13"]; Default: "tlsv1_2"
- cert String
- the path of ssl cert to use for the pgvector database
- cert
Key String - the path of ssl cert key to use for the pgvector database
- enabled Boolean
- whether to use ssl for the pgvector database. Default: true
- required Boolean
- whether ssl is required for the pgvector database. Default: true
- verify Boolean
- whether to verify ssl for the pgvector database. Default: true
- version String
- the ssl version to use for the pgvector database. possible known values include one of ["any", "tlsv12", "tlsv13"]; Default: "tlsv1_2"
- cert string
- the path of ssl cert to use for the pgvector database
- cert
Key string - the path of ssl cert key to use for the pgvector database
- enabled boolean
- whether to use ssl for the pgvector database. Default: true
- required boolean
- whether ssl is required for the pgvector database. Default: true
- verify boolean
- whether to verify ssl for the pgvector database. Default: true
- version string
- the ssl version to use for the pgvector database. possible known values include one of ["any", "tlsv12", "tlsv13"]; Default: "tlsv1_2"
- cert str
- the path of ssl cert to use for the pgvector database
- cert_
key str - the path of ssl cert key to use for the pgvector database
- enabled bool
- whether to use ssl for the pgvector database. Default: true
- required bool
- whether ssl is required for the pgvector database. Default: true
- verify bool
- whether to verify ssl for the pgvector database. Default: true
- version str
- the ssl version to use for the pgvector database. possible known values include one of ["any", "tlsv12", "tlsv13"]; Default: "tlsv1_2"
- cert String
- the path of ssl cert to use for the pgvector database
- cert
Key String - the path of ssl cert key to use for the pgvector database
- enabled Boolean
- whether to use ssl for the pgvector database. Default: true
- required Boolean
- whether ssl is required for the pgvector database. Default: true
- verify Boolean
- whether to verify ssl for the pgvector database. Default: true
- version String
- the ssl version to use for the pgvector database. possible known values include one of ["any", "tlsv12", "tlsv13"]; Default: "tlsv1_2"
AiGatewayModelModelConfigBalancerSemanticVectordbRedis, AiGatewayModelModelConfigBalancerSemanticVectordbRedisArgs
- Cloud
Authentication AiGateway Model Model Config Balancer Semantic Vectordb Redis Cloud Authentication - Auth related config for connecting to a Cloud Provider's Redis instance.
- Cluster
Ai
Gateway Model Model Config Balancer Semantic Vectordb Redis Cluster - Cluster configuration for the Redis connection.
- Connect
Timeout double - An integer representing a timeout in milliseconds. Must be between 0 and 2^31-2. Default: 2000
- Connection
Is boolProxied - If the connection to Redis is proxied (e.g. Envoy), set it
true. Set thehostandportto point to the proxy address. Default: false - Database double
- Database to use for the Redis connection when using the
redisstrategy. Default: 0 - Dimensions double
- the desired dimensionality for the vectors. Not Null
- Distance
Metric string - the distance metric to use for vector searches. possible known values include one of ["cosine", "euclidean"]; Not Null
- Host string
- A string representing a host name, such as example.com. This field is referenceable. Default: "127.0.0.1"
- Keepalive
Ai
Gateway Model Model Config Balancer Semantic Vectordb Redis Keepalive - Keepalive configuration for the Redis connection.
- Password string
- Password to use for Redis connections. If undefined, no AUTH commands are sent to Redis. This field is referenceable.
- Port string
- An integer representing a port number between 0 and 65535, inclusive. This field is referenceable. Default: "6379"
- Read
Timeout double - An integer representing a timeout in milliseconds. Must be between 0 and 2^31-2. Default: 2000
- Send
Timeout double - An integer representing a timeout in milliseconds. Must be between 0 and 2^31-2. Default: 2000
- Sentinel
Ai
Gateway Model Model Config Balancer Semantic Vectordb Redis Sentinel - Configuration for Redis Sentinel.
- Server
Name string - A string representing an SNI (server name indication) value for TLS. This field is referenceable.
- Ssl bool
- If set to true, uses SSL to connect to Redis. Default: true
- Ssl
Verify bool - If set to true, verifies the validity of the server SSL certificate. If setting this parameter, also configure
lua_ssl_trusted_certificateinkong.confto specify the CA (or server) certificate used by your Redis server. You may also need to configurelua_ssl_verify_depthaccordingly. Default: true - Threshold double
- the default similarity threshold for accepting semantic search results (float). Higher threshold means more results are considered similar.
- Username string
- Username to use for Redis connections. If undefined, ACL authentication won't be performed. This requires Redis v6.0.0+. To be compatible with Redis v5.x.y, you can set it to
default. This field is referenceable.
- Cloud
Authentication AiGateway Model Model Config Balancer Semantic Vectordb Redis Cloud Authentication - Auth related config for connecting to a Cloud Provider's Redis instance.
- Cluster
Ai
Gateway Model Model Config Balancer Semantic Vectordb Redis Cluster - Cluster configuration for the Redis connection.
- Connect
Timeout float64 - An integer representing a timeout in milliseconds. Must be between 0 and 2^31-2. Default: 2000
- Connection
Is boolProxied - If the connection to Redis is proxied (e.g. Envoy), set it
true. Set thehostandportto point to the proxy address. Default: false - Database float64
- Database to use for the Redis connection when using the
redisstrategy. Default: 0 - Dimensions float64
- the desired dimensionality for the vectors. Not Null
- Distance
Metric string - the distance metric to use for vector searches. possible known values include one of ["cosine", "euclidean"]; Not Null
- Host string
- A string representing a host name, such as example.com. This field is referenceable. Default: "127.0.0.1"
- Keepalive
Ai
Gateway Model Model Config Balancer Semantic Vectordb Redis Keepalive - Keepalive configuration for the Redis connection.
- Password string
- Password to use for Redis connections. If undefined, no AUTH commands are sent to Redis. This field is referenceable.
- Port string
- An integer representing a port number between 0 and 65535, inclusive. This field is referenceable. Default: "6379"
- Read
Timeout float64 - An integer representing a timeout in milliseconds. Must be between 0 and 2^31-2. Default: 2000
- Send
Timeout float64 - An integer representing a timeout in milliseconds. Must be between 0 and 2^31-2. Default: 2000
- Sentinel
Ai
Gateway Model Model Config Balancer Semantic Vectordb Redis Sentinel - Configuration for Redis Sentinel.
- Server
Name string - A string representing an SNI (server name indication) value for TLS. This field is referenceable.
- Ssl bool
- If set to true, uses SSL to connect to Redis. Default: true
- Ssl
Verify bool - If set to true, verifies the validity of the server SSL certificate. If setting this parameter, also configure
lua_ssl_trusted_certificateinkong.confto specify the CA (or server) certificate used by your Redis server. You may also need to configurelua_ssl_verify_depthaccordingly. Default: true - Threshold float64
- the default similarity threshold for accepting semantic search results (float). Higher threshold means more results are considered similar.
- Username string
- Username to use for Redis connections. If undefined, ACL authentication won't be performed. This requires Redis v6.0.0+. To be compatible with Redis v5.x.y, you can set it to
default. This field is referenceable.
- cloud_
authentication object - Auth related config for connecting to a Cloud Provider's Redis instance.
- cluster object
- Cluster configuration for the Redis connection.
- connect_
timeout number - An integer representing a timeout in milliseconds. Must be between 0 and 2^31-2. Default: 2000
- connection_
is_ boolproxied - If the connection to Redis is proxied (e.g. Envoy), set it
true. Set thehostandportto point to the proxy address. Default: false - database number
- Database to use for the Redis connection when using the
redisstrategy. Default: 0 - dimensions number
- the desired dimensionality for the vectors. Not Null
- distance_
metric string - the distance metric to use for vector searches. possible known values include one of ["cosine", "euclidean"]; Not Null
- host string
- A string representing a host name, such as example.com. This field is referenceable. Default: "127.0.0.1"
- keepalive object
- Keepalive configuration for the Redis connection.
- password string
- Password to use for Redis connections. If undefined, no AUTH commands are sent to Redis. This field is referenceable.
- port string
- An integer representing a port number between 0 and 65535, inclusive. This field is referenceable. Default: "6379"
- read_
timeout number - An integer representing a timeout in milliseconds. Must be between 0 and 2^31-2. Default: 2000
- send_
timeout number - An integer representing a timeout in milliseconds. Must be between 0 and 2^31-2. Default: 2000
- sentinel object
- Configuration for Redis Sentinel.
- server_
name string - A string representing an SNI (server name indication) value for TLS. This field is referenceable.
- ssl bool
- If set to true, uses SSL to connect to Redis. Default: true
- ssl_
verify bool - If set to true, verifies the validity of the server SSL certificate. If setting this parameter, also configure
lua_ssl_trusted_certificateinkong.confto specify the CA (or server) certificate used by your Redis server. You may also need to configurelua_ssl_verify_depthaccordingly. Default: true - threshold number
- the default similarity threshold for accepting semantic search results (float). Higher threshold means more results are considered similar.
- username string
- Username to use for Redis connections. If undefined, ACL authentication won't be performed. This requires Redis v6.0.0+. To be compatible with Redis v5.x.y, you can set it to
default. This field is referenceable.
- cloud
Authentication AiGateway Model Model Config Balancer Semantic Vectordb Redis Cloud Authentication - Auth related config for connecting to a Cloud Provider's Redis instance.
- cluster
Ai
Gateway Model Model Config Balancer Semantic Vectordb Redis Cluster - Cluster configuration for the Redis connection.
- connect
Timeout Double - An integer representing a timeout in milliseconds. Must be between 0 and 2^31-2. Default: 2000
- connection
Is BooleanProxied - If the connection to Redis is proxied (e.g. Envoy), set it
true. Set thehostandportto point to the proxy address. Default: false - database Double
- Database to use for the Redis connection when using the
redisstrategy. Default: 0 - dimensions Double
- the desired dimensionality for the vectors. Not Null
- distance
Metric String - the distance metric to use for vector searches. possible known values include one of ["cosine", "euclidean"]; Not Null
- host String
- A string representing a host name, such as example.com. This field is referenceable. Default: "127.0.0.1"
- keepalive
Ai
Gateway Model Model Config Balancer Semantic Vectordb Redis Keepalive - Keepalive configuration for the Redis connection.
- password String
- Password to use for Redis connections. If undefined, no AUTH commands are sent to Redis. This field is referenceable.
- port String
- An integer representing a port number between 0 and 65535, inclusive. This field is referenceable. Default: "6379"
- read
Timeout Double - An integer representing a timeout in milliseconds. Must be between 0 and 2^31-2. Default: 2000
- send
Timeout Double - An integer representing a timeout in milliseconds. Must be between 0 and 2^31-2. Default: 2000
- sentinel
Ai
Gateway Model Model Config Balancer Semantic Vectordb Redis Sentinel - Configuration for Redis Sentinel.
- server
Name String - A string representing an SNI (server name indication) value for TLS. This field is referenceable.
- ssl Boolean
- If set to true, uses SSL to connect to Redis. Default: true
- ssl
Verify Boolean - If set to true, verifies the validity of the server SSL certificate. If setting this parameter, also configure
lua_ssl_trusted_certificateinkong.confto specify the CA (or server) certificate used by your Redis server. You may also need to configurelua_ssl_verify_depthaccordingly. Default: true - threshold Double
- the default similarity threshold for accepting semantic search results (float). Higher threshold means more results are considered similar.
- username String
- Username to use for Redis connections. If undefined, ACL authentication won't be performed. This requires Redis v6.0.0+. To be compatible with Redis v5.x.y, you can set it to
default. This field is referenceable.
- cloud
Authentication AiGateway Model Model Config Balancer Semantic Vectordb Redis Cloud Authentication - Auth related config for connecting to a Cloud Provider's Redis instance.
- cluster
Ai
Gateway Model Model Config Balancer Semantic Vectordb Redis Cluster - Cluster configuration for the Redis connection.
- connect
Timeout number - An integer representing a timeout in milliseconds. Must be between 0 and 2^31-2. Default: 2000
- connection
Is booleanProxied - If the connection to Redis is proxied (e.g. Envoy), set it
true. Set thehostandportto point to the proxy address. Default: false - database number
- Database to use for the Redis connection when using the
redisstrategy. Default: 0 - dimensions number
- the desired dimensionality for the vectors. Not Null
- distance
Metric string - the distance metric to use for vector searches. possible known values include one of ["cosine", "euclidean"]; Not Null
- host string
- A string representing a host name, such as example.com. This field is referenceable. Default: "127.0.0.1"
- keepalive
Ai
Gateway Model Model Config Balancer Semantic Vectordb Redis Keepalive - Keepalive configuration for the Redis connection.
- password string
- Password to use for Redis connections. If undefined, no AUTH commands are sent to Redis. This field is referenceable.
- port string
- An integer representing a port number between 0 and 65535, inclusive. This field is referenceable. Default: "6379"
- read
Timeout number - An integer representing a timeout in milliseconds. Must be between 0 and 2^31-2. Default: 2000
- send
Timeout number - An integer representing a timeout in milliseconds. Must be between 0 and 2^31-2. Default: 2000
- sentinel
Ai
Gateway Model Model Config Balancer Semantic Vectordb Redis Sentinel - Configuration for Redis Sentinel.
- server
Name string - A string representing an SNI (server name indication) value for TLS. This field is referenceable.
- ssl boolean
- If set to true, uses SSL to connect to Redis. Default: true
- ssl
Verify boolean - If set to true, verifies the validity of the server SSL certificate. If setting this parameter, also configure
lua_ssl_trusted_certificateinkong.confto specify the CA (or server) certificate used by your Redis server. You may also need to configurelua_ssl_verify_depthaccordingly. Default: true - threshold number
- the default similarity threshold for accepting semantic search results (float). Higher threshold means more results are considered similar.
- username string
- Username to use for Redis connections. If undefined, ACL authentication won't be performed. This requires Redis v6.0.0+. To be compatible with Redis v5.x.y, you can set it to
default. This field is referenceable.
- cloud_
authentication AiGateway Model Model Config Balancer Semantic Vectordb Redis Cloud Authentication - Auth related config for connecting to a Cloud Provider's Redis instance.
- cluster
Ai
Gateway Model Model Config Balancer Semantic Vectordb Redis Cluster - Cluster configuration for the Redis connection.
- connect_
timeout float - An integer representing a timeout in milliseconds. Must be between 0 and 2^31-2. Default: 2000
- connection_
is_ boolproxied - If the connection to Redis is proxied (e.g. Envoy), set it
true. Set thehostandportto point to the proxy address. Default: false - database float
- Database to use for the Redis connection when using the
redisstrategy. Default: 0 - dimensions float
- the desired dimensionality for the vectors. Not Null
- distance_
metric str - the distance metric to use for vector searches. possible known values include one of ["cosine", "euclidean"]; Not Null
- host str
- A string representing a host name, such as example.com. This field is referenceable. Default: "127.0.0.1"
- keepalive
Ai
Gateway Model Model Config Balancer Semantic Vectordb Redis Keepalive - Keepalive configuration for the Redis connection.
- password str
- Password to use for Redis connections. If undefined, no AUTH commands are sent to Redis. This field is referenceable.
- port str
- An integer representing a port number between 0 and 65535, inclusive. This field is referenceable. Default: "6379"
- read_
timeout float - An integer representing a timeout in milliseconds. Must be between 0 and 2^31-2. Default: 2000
- send_
timeout float - An integer representing a timeout in milliseconds. Must be between 0 and 2^31-2. Default: 2000
- sentinel
Ai
Gateway Model Model Config Balancer Semantic Vectordb Redis Sentinel - Configuration for Redis Sentinel.
- server_
name str - A string representing an SNI (server name indication) value for TLS. This field is referenceable.
- ssl bool
- If set to true, uses SSL to connect to Redis. Default: true
- ssl_
verify bool - If set to true, verifies the validity of the server SSL certificate. If setting this parameter, also configure
lua_ssl_trusted_certificateinkong.confto specify the CA (or server) certificate used by your Redis server. You may also need to configurelua_ssl_verify_depthaccordingly. Default: true - threshold float
- the default similarity threshold for accepting semantic search results (float). Higher threshold means more results are considered similar.
- username str
- Username to use for Redis connections. If undefined, ACL authentication won't be performed. This requires Redis v6.0.0+. To be compatible with Redis v5.x.y, you can set it to
default. This field is referenceable.
- cloud
Authentication Property Map - Auth related config for connecting to a Cloud Provider's Redis instance.
- cluster Property Map
- Cluster configuration for the Redis connection.
- connect
Timeout Number - An integer representing a timeout in milliseconds. Must be between 0 and 2^31-2. Default: 2000
- connection
Is BooleanProxied - If the connection to Redis is proxied (e.g. Envoy), set it
true. Set thehostandportto point to the proxy address. Default: false - database Number
- Database to use for the Redis connection when using the
redisstrategy. Default: 0 - dimensions Number
- the desired dimensionality for the vectors. Not Null
- distance
Metric String - the distance metric to use for vector searches. possible known values include one of ["cosine", "euclidean"]; Not Null
- host String
- A string representing a host name, such as example.com. This field is referenceable. Default: "127.0.0.1"
- keepalive Property Map
- Keepalive configuration for the Redis connection.
- password String
- Password to use for Redis connections. If undefined, no AUTH commands are sent to Redis. This field is referenceable.
- port String
- An integer representing a port number between 0 and 65535, inclusive. This field is referenceable. Default: "6379"
- read
Timeout Number - An integer representing a timeout in milliseconds. Must be between 0 and 2^31-2. Default: 2000
- send
Timeout Number - An integer representing a timeout in milliseconds. Must be between 0 and 2^31-2. Default: 2000
- sentinel Property Map
- Configuration for Redis Sentinel.
- server
Name String - A string representing an SNI (server name indication) value for TLS. This field is referenceable.
- ssl Boolean
- If set to true, uses SSL to connect to Redis. Default: true
- ssl
Verify Boolean - If set to true, verifies the validity of the server SSL certificate. If setting this parameter, also configure
lua_ssl_trusted_certificateinkong.confto specify the CA (or server) certificate used by your Redis server. You may also need to configurelua_ssl_verify_depthaccordingly. Default: true - threshold Number
- the default similarity threshold for accepting semantic search results (float). Higher threshold means more results are considered similar.
- username String
- Username to use for Redis connections. If undefined, ACL authentication won't be performed. This requires Redis v6.0.0+. To be compatible with Redis v5.x.y, you can set it to
default. This field is referenceable.
AiGatewayModelModelConfigBalancerSemanticVectordbRedisCloudAuthentication, AiGatewayModelModelConfigBalancerSemanticVectordbRedisCloudAuthenticationArgs
- Aws
Ai
Gateway Model Model Config Balancer Semantic Vectordb Redis Cloud Authentication Aws - AWS specific configs for connecting to a Cloud Provider's redis instance.
- Azure
Ai
Gateway Model Model Config Balancer Semantic Vectordb Redis Cloud Authentication Azure - Azure specific configs for connecting to a Cloud Provider's redis instance.
- Gcp
Ai
Gateway Model Model Config Balancer Semantic Vectordb Redis Cloud Authentication Gcp - GCP specific configs for connecting to a Cloud Provider's redis instance.
- Aws
Ai
Gateway Model Model Config Balancer Semantic Vectordb Redis Cloud Authentication Aws - AWS specific configs for connecting to a Cloud Provider's redis instance.
- Azure
Ai
Gateway Model Model Config Balancer Semantic Vectordb Redis Cloud Authentication Azure - Azure specific configs for connecting to a Cloud Provider's redis instance.
- Gcp
Ai
Gateway Model Model Config Balancer Semantic Vectordb Redis Cloud Authentication Gcp - GCP specific configs for connecting to a Cloud Provider's redis instance.
- aws
Ai
Gateway Model Model Config Balancer Semantic Vectordb Redis Cloud Authentication Aws - AWS specific configs for connecting to a Cloud Provider's redis instance.
- azure
Ai
Gateway Model Model Config Balancer Semantic Vectordb Redis Cloud Authentication Azure - Azure specific configs for connecting to a Cloud Provider's redis instance.
- gcp
Ai
Gateway Model Model Config Balancer Semantic Vectordb Redis Cloud Authentication Gcp - GCP specific configs for connecting to a Cloud Provider's redis instance.
- aws
Ai
Gateway Model Model Config Balancer Semantic Vectordb Redis Cloud Authentication Aws - AWS specific configs for connecting to a Cloud Provider's redis instance.
- azure
Ai
Gateway Model Model Config Balancer Semantic Vectordb Redis Cloud Authentication Azure - Azure specific configs for connecting to a Cloud Provider's redis instance.
- gcp
Ai
Gateway Model Model Config Balancer Semantic Vectordb Redis Cloud Authentication Gcp - GCP specific configs for connecting to a Cloud Provider's redis instance.
- aws
Ai
Gateway Model Model Config Balancer Semantic Vectordb Redis Cloud Authentication Aws - AWS specific configs for connecting to a Cloud Provider's redis instance.
- azure
Ai
Gateway Model Model Config Balancer Semantic Vectordb Redis Cloud Authentication Azure - Azure specific configs for connecting to a Cloud Provider's redis instance.
- gcp
Ai
Gateway Model Model Config Balancer Semantic Vectordb Redis Cloud Authentication Gcp - GCP specific configs for connecting to a Cloud Provider's redis instance.
- aws Property Map
- AWS specific configs for connecting to a Cloud Provider's redis instance.
- azure Property Map
- Azure specific configs for connecting to a Cloud Provider's redis instance.
- gcp Property Map
- GCP specific configs for connecting to a Cloud Provider's redis instance.
AiGatewayModelModelConfigBalancerSemanticVectordbRedisCloudAuthenticationAws, AiGatewayModelModelConfigBalancerSemanticVectordbRedisCloudAuthenticationAwsArgs
- Access
Key stringId - AWS Access Key ID to be used for authentication. This field is referenceable.
- Assume
Role stringArn - The ARN of the IAM role to assume for generating ElastiCache IAM authentication tokens. This field is referenceable.
- Cache
Name string - The name of the AWS Elasticache cluster. This field is referenceable.
- Is
Serverless bool - This flag specifies whether the cluster is serverless. Default: true
- Region string
- The region of the AWS ElastiCache cluster. This field is referenceable.
- Role
Session stringName - The session name for the temporary credentials when assuming the IAM role. This field is referenceable.
- Secret
Access stringKey - AWS Secret Access Key. This field is referenceable.
- Access
Key stringId - AWS Access Key ID to be used for authentication. This field is referenceable.
- Assume
Role stringArn - The ARN of the IAM role to assume for generating ElastiCache IAM authentication tokens. This field is referenceable.
- Cache
Name string - The name of the AWS Elasticache cluster. This field is referenceable.
- Is
Serverless bool - This flag specifies whether the cluster is serverless. Default: true
- Region string
- The region of the AWS ElastiCache cluster. This field is referenceable.
- Role
Session stringName - The session name for the temporary credentials when assuming the IAM role. This field is referenceable.
- Secret
Access stringKey - AWS Secret Access Key. This field is referenceable.
- access_
key_ stringid - AWS Access Key ID to be used for authentication. This field is referenceable.
- assume_
role_ stringarn - The ARN of the IAM role to assume for generating ElastiCache IAM authentication tokens. This field is referenceable.
- cache_
name string - The name of the AWS Elasticache cluster. This field is referenceable.
- is_
serverless bool - This flag specifies whether the cluster is serverless. Default: true
- region string
- The region of the AWS ElastiCache cluster. This field is referenceable.
- role_
session_ stringname - The session name for the temporary credentials when assuming the IAM role. This field is referenceable.
- secret_
access_ stringkey - AWS Secret Access Key. This field is referenceable.
- access
Key StringId - AWS Access Key ID to be used for authentication. This field is referenceable.
- assume
Role StringArn - The ARN of the IAM role to assume for generating ElastiCache IAM authentication tokens. This field is referenceable.
- cache
Name String - The name of the AWS Elasticache cluster. This field is referenceable.
- is
Serverless Boolean - This flag specifies whether the cluster is serverless. Default: true
- region String
- The region of the AWS ElastiCache cluster. This field is referenceable.
- role
Session StringName - The session name for the temporary credentials when assuming the IAM role. This field is referenceable.
- secret
Access StringKey - AWS Secret Access Key. This field is referenceable.
- access
Key stringId - AWS Access Key ID to be used for authentication. This field is referenceable.
- assume
Role stringArn - The ARN of the IAM role to assume for generating ElastiCache IAM authentication tokens. This field is referenceable.
- cache
Name string - The name of the AWS Elasticache cluster. This field is referenceable.
- is
Serverless boolean - This flag specifies whether the cluster is serverless. Default: true
- region string
- The region of the AWS ElastiCache cluster. This field is referenceable.
- role
Session stringName - The session name for the temporary credentials when assuming the IAM role. This field is referenceable.
- secret
Access stringKey - AWS Secret Access Key. This field is referenceable.
- access_
key_ strid - AWS Access Key ID to be used for authentication. This field is referenceable.
- assume_
role_ strarn - The ARN of the IAM role to assume for generating ElastiCache IAM authentication tokens. This field is referenceable.
- cache_
name str - The name of the AWS Elasticache cluster. This field is referenceable.
- is_
serverless bool - This flag specifies whether the cluster is serverless. Default: true
- region str
- The region of the AWS ElastiCache cluster. This field is referenceable.
- role_
session_ strname - The session name for the temporary credentials when assuming the IAM role. This field is referenceable.
- secret_
access_ strkey - AWS Secret Access Key. This field is referenceable.
- access
Key StringId - AWS Access Key ID to be used for authentication. This field is referenceable.
- assume
Role StringArn - The ARN of the IAM role to assume for generating ElastiCache IAM authentication tokens. This field is referenceable.
- cache
Name String - The name of the AWS Elasticache cluster. This field is referenceable.
- is
Serverless Boolean - This flag specifies whether the cluster is serverless. Default: true
- region String
- The region of the AWS ElastiCache cluster. This field is referenceable.
- role
Session StringName - The session name for the temporary credentials when assuming the IAM role. This field is referenceable.
- secret
Access StringKey - AWS Secret Access Key. This field is referenceable.
AiGatewayModelModelConfigBalancerSemanticVectordbRedisCloudAuthenticationAzure, AiGatewayModelModelConfigBalancerSemanticVectordbRedisCloudAuthenticationAzureArgs
- Client
Id string - Azure Client ID. This field is referenceable.
- Client
Secret string - Azure Client Secret. This field is referenceable.
- Tenant
Id string - Azure Tenant ID. This field is referenceable.
- Client
Id string - Azure Client ID. This field is referenceable.
- Client
Secret string - Azure Client Secret. This field is referenceable.
- Tenant
Id string - Azure Tenant ID. This field is referenceable.
- client_
id string - Azure Client ID. This field is referenceable.
- client_
secret string - Azure Client Secret. This field is referenceable.
- tenant_
id string - Azure Tenant ID. This field is referenceable.
- client
Id String - Azure Client ID. This field is referenceable.
- client
Secret String - Azure Client Secret. This field is referenceable.
- tenant
Id String - Azure Tenant ID. This field is referenceable.
- client
Id string - Azure Client ID. This field is referenceable.
- client
Secret string - Azure Client Secret. This field is referenceable.
- tenant
Id string - Azure Tenant ID. This field is referenceable.
- client_
id str - Azure Client ID. This field is referenceable.
- client_
secret str - Azure Client Secret. This field is referenceable.
- tenant_
id str - Azure Tenant ID. This field is referenceable.
- client
Id String - Azure Client ID. This field is referenceable.
- client
Secret String - Azure Client Secret. This field is referenceable.
- tenant
Id String - Azure Tenant ID. This field is referenceable.
AiGatewayModelModelConfigBalancerSemanticVectordbRedisCloudAuthenticationGcp, AiGatewayModelModelConfigBalancerSemanticVectordbRedisCloudAuthenticationGcpArgs
- Service
Account stringJson - GCP Service Account JSON. This field is referenceable.
- Service
Account stringJson - GCP Service Account JSON. This field is referenceable.
- service_
account_ stringjson - GCP Service Account JSON. This field is referenceable.
- service
Account StringJson - GCP Service Account JSON. This field is referenceable.
- service
Account stringJson - GCP Service Account JSON. This field is referenceable.
- service_
account_ strjson - GCP Service Account JSON. This field is referenceable.
- service
Account StringJson - GCP Service Account JSON. This field is referenceable.
AiGatewayModelModelConfigBalancerSemanticVectordbRedisCluster, AiGatewayModelModelConfigBalancerSemanticVectordbRedisClusterArgs
- Max
Redirections double - Maximum retry attempts for redirection. Default: 5
- Nodes
List<Ai
Gateway Model Model Config Balancer Semantic Vectordb Redis Cluster Node> - Cluster addresses to use for Redis connections when the
redisstrategy is defined. Defining this field implies using a Redis Cluster. The minimum length of the array is 1 element.
- Max
Redirections float64 - Maximum retry attempts for redirection. Default: 5
- Nodes
[]Ai
Gateway Model Model Config Balancer Semantic Vectordb Redis Cluster Node - Cluster addresses to use for Redis connections when the
redisstrategy is defined. Defining this field implies using a Redis Cluster. The minimum length of the array is 1 element.
- max_
redirections number - Maximum retry attempts for redirection. Default: 5
- nodes list(object)
- Cluster addresses to use for Redis connections when the
redisstrategy is defined. Defining this field implies using a Redis Cluster. The minimum length of the array is 1 element.
- max
Redirections Double - Maximum retry attempts for redirection. Default: 5
- nodes
List<Ai
Gateway Model Model Config Balancer Semantic Vectordb Redis Cluster Node> - Cluster addresses to use for Redis connections when the
redisstrategy is defined. Defining this field implies using a Redis Cluster. The minimum length of the array is 1 element.
- max
Redirections number - Maximum retry attempts for redirection. Default: 5
- nodes
Ai
Gateway Model Model Config Balancer Semantic Vectordb Redis Cluster Node[] - Cluster addresses to use for Redis connections when the
redisstrategy is defined. Defining this field implies using a Redis Cluster. The minimum length of the array is 1 element.
- max_
redirections float - Maximum retry attempts for redirection. Default: 5
- nodes
Sequence[Ai
Gateway Model Model Config Balancer Semantic Vectordb Redis Cluster Node] - Cluster addresses to use for Redis connections when the
redisstrategy is defined. Defining this field implies using a Redis Cluster. The minimum length of the array is 1 element.
- max
Redirections Number - Maximum retry attempts for redirection. Default: 5
- nodes List<Property Map>
- Cluster addresses to use for Redis connections when the
redisstrategy is defined. Defining this field implies using a Redis Cluster. The minimum length of the array is 1 element.
AiGatewayModelModelConfigBalancerSemanticVectordbRedisClusterNode, AiGatewayModelModelConfigBalancerSemanticVectordbRedisClusterNodeArgs
AiGatewayModelModelConfigBalancerSemanticVectordbRedisKeepalive, AiGatewayModelModelConfigBalancerSemanticVectordbRedisKeepaliveArgs
- Backlog double
- Limits the total number of opened connections for a pool. If the connection pool is full, connection queues above the limit go into the backlog queue. If the backlog queue is full, subsequent connect operations fail and return
nil. Queued operations (subject to set timeouts) resume once the number of connections in the pool is less thanpool_size. If latency is high or throughput is low, try increasing this value. Empirically, this value is larger thanpool_size. - Pool
Size double - The size limit for every cosocket connection pool associated with every remote server, per worker process. If neither
pool_sizenorbacklogis specified, no pool is created. Ifpool_sizeisn't specified butbacklogis specified, then the pool uses the default value. Try to increase (e.g. 512) this value if latency is high or throughput is low. Default: 256
- Backlog float64
- Limits the total number of opened connections for a pool. If the connection pool is full, connection queues above the limit go into the backlog queue. If the backlog queue is full, subsequent connect operations fail and return
nil. Queued operations (subject to set timeouts) resume once the number of connections in the pool is less thanpool_size. If latency is high or throughput is low, try increasing this value. Empirically, this value is larger thanpool_size. - Pool
Size float64 - The size limit for every cosocket connection pool associated with every remote server, per worker process. If neither
pool_sizenorbacklogis specified, no pool is created. Ifpool_sizeisn't specified butbacklogis specified, then the pool uses the default value. Try to increase (e.g. 512) this value if latency is high or throughput is low. Default: 256
- backlog number
- Limits the total number of opened connections for a pool. If the connection pool is full, connection queues above the limit go into the backlog queue. If the backlog queue is full, subsequent connect operations fail and return
nil. Queued operations (subject to set timeouts) resume once the number of connections in the pool is less thanpool_size. If latency is high or throughput is low, try increasing this value. Empirically, this value is larger thanpool_size. - pool_
size number - The size limit for every cosocket connection pool associated with every remote server, per worker process. If neither
pool_sizenorbacklogis specified, no pool is created. Ifpool_sizeisn't specified butbacklogis specified, then the pool uses the default value. Try to increase (e.g. 512) this value if latency is high or throughput is low. Default: 256
- backlog Double
- Limits the total number of opened connections for a pool. If the connection pool is full, connection queues above the limit go into the backlog queue. If the backlog queue is full, subsequent connect operations fail and return
nil. Queued operations (subject to set timeouts) resume once the number of connections in the pool is less thanpool_size. If latency is high or throughput is low, try increasing this value. Empirically, this value is larger thanpool_size. - pool
Size Double - The size limit for every cosocket connection pool associated with every remote server, per worker process. If neither
pool_sizenorbacklogis specified, no pool is created. Ifpool_sizeisn't specified butbacklogis specified, then the pool uses the default value. Try to increase (e.g. 512) this value if latency is high or throughput is low. Default: 256
- backlog number
- Limits the total number of opened connections for a pool. If the connection pool is full, connection queues above the limit go into the backlog queue. If the backlog queue is full, subsequent connect operations fail and return
nil. Queued operations (subject to set timeouts) resume once the number of connections in the pool is less thanpool_size. If latency is high or throughput is low, try increasing this value. Empirically, this value is larger thanpool_size. - pool
Size number - The size limit for every cosocket connection pool associated with every remote server, per worker process. If neither
pool_sizenorbacklogis specified, no pool is created. Ifpool_sizeisn't specified butbacklogis specified, then the pool uses the default value. Try to increase (e.g. 512) this value if latency is high or throughput is low. Default: 256
- backlog float
- Limits the total number of opened connections for a pool. If the connection pool is full, connection queues above the limit go into the backlog queue. If the backlog queue is full, subsequent connect operations fail and return
nil. Queued operations (subject to set timeouts) resume once the number of connections in the pool is less thanpool_size. If latency is high or throughput is low, try increasing this value. Empirically, this value is larger thanpool_size. - pool_
size float - The size limit for every cosocket connection pool associated with every remote server, per worker process. If neither
pool_sizenorbacklogis specified, no pool is created. Ifpool_sizeisn't specified butbacklogis specified, then the pool uses the default value. Try to increase (e.g. 512) this value if latency is high or throughput is low. Default: 256
- backlog Number
- Limits the total number of opened connections for a pool. If the connection pool is full, connection queues above the limit go into the backlog queue. If the backlog queue is full, subsequent connect operations fail and return
nil. Queued operations (subject to set timeouts) resume once the number of connections in the pool is less thanpool_size. If latency is high or throughput is low, try increasing this value. Empirically, this value is larger thanpool_size. - pool
Size Number - The size limit for every cosocket connection pool associated with every remote server, per worker process. If neither
pool_sizenorbacklogis specified, no pool is created. Ifpool_sizeisn't specified butbacklogis specified, then the pool uses the default value. Try to increase (e.g. 512) this value if latency is high or throughput is low. Default: 256
AiGatewayModelModelConfigBalancerSemanticVectordbRedisSentinel, AiGatewayModelModelConfigBalancerSemanticVectordbRedisSentinelArgs
- Master string
- Sentinel master to use for Redis connections. Defining this value implies using Redis Sentinel.
- Nodes
List<Ai
Gateway Model Model Config Balancer Semantic Vectordb Redis Sentinel Node> - Sentinel node addresses to use for Redis connections when the
redisstrategy is defined. Defining this field implies using a Redis Sentinel. The minimum length of the array is 1 element. - Password string
- Sentinel password to authenticate with a Redis Sentinel instance. If undefined, no AUTH commands are sent to Redis Sentinels. This field is referenceable.
- Role string
- Sentinel role to use for Redis connections when the
redisstrategy is defined. Defining this value implies using Redis Sentinel. possible known values include one of ["any", "master", "slave"] - Username string
- Sentinel username to authenticate with a Redis Sentinel instance. If undefined, ACL authentication won't be performed. This requires Redis v6.2.0+. This field is referenceable.
- Master string
- Sentinel master to use for Redis connections. Defining this value implies using Redis Sentinel.
- Nodes
[]Ai
Gateway Model Model Config Balancer Semantic Vectordb Redis Sentinel Node - Sentinel node addresses to use for Redis connections when the
redisstrategy is defined. Defining this field implies using a Redis Sentinel. The minimum length of the array is 1 element. - Password string
- Sentinel password to authenticate with a Redis Sentinel instance. If undefined, no AUTH commands are sent to Redis Sentinels. This field is referenceable.
- Role string
- Sentinel role to use for Redis connections when the
redisstrategy is defined. Defining this value implies using Redis Sentinel. possible known values include one of ["any", "master", "slave"] - Username string
- Sentinel username to authenticate with a Redis Sentinel instance. If undefined, ACL authentication won't be performed. This requires Redis v6.2.0+. This field is referenceable.
- master string
- Sentinel master to use for Redis connections. Defining this value implies using Redis Sentinel.
- nodes list(object)
- Sentinel node addresses to use for Redis connections when the
redisstrategy is defined. Defining this field implies using a Redis Sentinel. The minimum length of the array is 1 element. - password string
- Sentinel password to authenticate with a Redis Sentinel instance. If undefined, no AUTH commands are sent to Redis Sentinels. This field is referenceable.
- role string
- Sentinel role to use for Redis connections when the
redisstrategy is defined. Defining this value implies using Redis Sentinel. possible known values include one of ["any", "master", "slave"] - username string
- Sentinel username to authenticate with a Redis Sentinel instance. If undefined, ACL authentication won't be performed. This requires Redis v6.2.0+. This field is referenceable.
- master String
- Sentinel master to use for Redis connections. Defining this value implies using Redis Sentinel.
- nodes
List<Ai
Gateway Model Model Config Balancer Semantic Vectordb Redis Sentinel Node> - Sentinel node addresses to use for Redis connections when the
redisstrategy is defined. Defining this field implies using a Redis Sentinel. The minimum length of the array is 1 element. - password String
- Sentinel password to authenticate with a Redis Sentinel instance. If undefined, no AUTH commands are sent to Redis Sentinels. This field is referenceable.
- role String
- Sentinel role to use for Redis connections when the
redisstrategy is defined. Defining this value implies using Redis Sentinel. possible known values include one of ["any", "master", "slave"] - username String
- Sentinel username to authenticate with a Redis Sentinel instance. If undefined, ACL authentication won't be performed. This requires Redis v6.2.0+. This field is referenceable.
- master string
- Sentinel master to use for Redis connections. Defining this value implies using Redis Sentinel.
- nodes
Ai
Gateway Model Model Config Balancer Semantic Vectordb Redis Sentinel Node[] - Sentinel node addresses to use for Redis connections when the
redisstrategy is defined. Defining this field implies using a Redis Sentinel. The minimum length of the array is 1 element. - password string
- Sentinel password to authenticate with a Redis Sentinel instance. If undefined, no AUTH commands are sent to Redis Sentinels. This field is referenceable.
- role string
- Sentinel role to use for Redis connections when the
redisstrategy is defined. Defining this value implies using Redis Sentinel. possible known values include one of ["any", "master", "slave"] - username string
- Sentinel username to authenticate with a Redis Sentinel instance. If undefined, ACL authentication won't be performed. This requires Redis v6.2.0+. This field is referenceable.
- master str
- Sentinel master to use for Redis connections. Defining this value implies using Redis Sentinel.
- nodes
Sequence[Ai
Gateway Model Model Config Balancer Semantic Vectordb Redis Sentinel Node] - Sentinel node addresses to use for Redis connections when the
redisstrategy is defined. Defining this field implies using a Redis Sentinel. The minimum length of the array is 1 element. - password str
- Sentinel password to authenticate with a Redis Sentinel instance. If undefined, no AUTH commands are sent to Redis Sentinels. This field is referenceable.
- role str
- Sentinel role to use for Redis connections when the
redisstrategy is defined. Defining this value implies using Redis Sentinel. possible known values include one of ["any", "master", "slave"] - username str
- Sentinel username to authenticate with a Redis Sentinel instance. If undefined, ACL authentication won't be performed. This requires Redis v6.2.0+. This field is referenceable.
- master String
- Sentinel master to use for Redis connections. Defining this value implies using Redis Sentinel.
- nodes List<Property Map>
- Sentinel node addresses to use for Redis connections when the
redisstrategy is defined. Defining this field implies using a Redis Sentinel. The minimum length of the array is 1 element. - password String
- Sentinel password to authenticate with a Redis Sentinel instance. If undefined, no AUTH commands are sent to Redis Sentinels. This field is referenceable.
- role String
- Sentinel role to use for Redis connections when the
redisstrategy is defined. Defining this value implies using Redis Sentinel. possible known values include one of ["any", "master", "slave"] - username String
- Sentinel username to authenticate with a Redis Sentinel instance. If undefined, ACL authentication won't be performed. This requires Redis v6.2.0+. This field is referenceable.
AiGatewayModelModelConfigBalancerSemanticVectordbRedisSentinelNode, AiGatewayModelModelConfigBalancerSemanticVectordbRedisSentinelNodeArgs
AiGatewayModelModelConfigLogging, AiGatewayModelModelConfigLoggingArgs
- Payloads bool
- Default: false
- Payloads bool
- Default: false
- payloads bool
- Default: false
- payloads Boolean
- Default: false
- payloads boolean
- Default: false
- payloads bool
- Default: false
- payloads Boolean
- Default: false
AiGatewayModelModelConfigModel, AiGatewayModelModelConfigModelArgs
- Name
Header bool - Display the model name selected in the X-Kong-LLM-Model response header. Default: true
- Name
Header bool - Display the model name selected in the X-Kong-LLM-Model response header. Default: true
- name_
header bool - Display the model name selected in the X-Kong-LLM-Model response header. Default: true
- name
Header Boolean - Display the model name selected in the X-Kong-LLM-Model response header. Default: true
- name
Header boolean - Display the model name selected in the X-Kong-LLM-Model response header. Default: true
- name_
header bool - Display the model name selected in the X-Kong-LLM-Model response header. Default: true
- name
Header Boolean - Display the model name selected in the X-Kong-LLM-Model response header. Default: true
AiGatewayModelModelConfigProxy, AiGatewayModelModelConfigProxyArgs
- Auth
Ai
Gateway Model Model Config Proxy Auth - Credentials used to authenticate to the proxy server.
- Http
Proxy AiGateway Model Model Config Proxy Http Proxy - HTTP proxy server to route plaintext outbound requests through.
- Https
Proxy AiGateway Model Model Config Proxy Https Proxy - HTTPS proxy server to route TLS outbound requests through.
- No
Proxy string - Comma-separated list of hosts that should not be proxied.
- Proxy
Scheme string - The proxy scheme to use when connecting to the proxy server. Default: "http"; must be "http"
- Auth
Ai
Gateway Model Model Config Proxy Auth - Credentials used to authenticate to the proxy server.
- Http
Proxy AiGateway Model Model Config Proxy Http Proxy - HTTP proxy server to route plaintext outbound requests through.
- Https
Proxy AiGateway Model Model Config Proxy Https Proxy - HTTPS proxy server to route TLS outbound requests through.
- No
Proxy string - Comma-separated list of hosts that should not be proxied.
- Proxy
Scheme string - The proxy scheme to use when connecting to the proxy server. Default: "http"; must be "http"
- auth object
- Credentials used to authenticate to the proxy server.
- http_
proxy object - HTTP proxy server to route plaintext outbound requests through.
- https_
proxy object - HTTPS proxy server to route TLS outbound requests through.
- no_
proxy string - Comma-separated list of hosts that should not be proxied.
- proxy_
scheme string - The proxy scheme to use when connecting to the proxy server. Default: "http"; must be "http"
- auth
Ai
Gateway Model Model Config Proxy Auth - Credentials used to authenticate to the proxy server.
- http
Proxy AiGateway Model Model Config Proxy Http Proxy - HTTP proxy server to route plaintext outbound requests through.
- https
Proxy AiGateway Model Model Config Proxy Https Proxy - HTTPS proxy server to route TLS outbound requests through.
- no
Proxy String - Comma-separated list of hosts that should not be proxied.
- proxy
Scheme String - The proxy scheme to use when connecting to the proxy server. Default: "http"; must be "http"
- auth
Ai
Gateway Model Model Config Proxy Auth - Credentials used to authenticate to the proxy server.
- http
Proxy AiGateway Model Model Config Proxy Http Proxy - HTTP proxy server to route plaintext outbound requests through.
- https
Proxy AiGateway Model Model Config Proxy Https Proxy - HTTPS proxy server to route TLS outbound requests through.
- no
Proxy string - Comma-separated list of hosts that should not be proxied.
- proxy
Scheme string - The proxy scheme to use when connecting to the proxy server. Default: "http"; must be "http"
- auth
Ai
Gateway Model Model Config Proxy Auth - Credentials used to authenticate to the proxy server.
- http_
proxy AiGateway Model Model Config Proxy Http Proxy - HTTP proxy server to route plaintext outbound requests through.
- https_
proxy AiGateway Model Model Config Proxy Https Proxy - HTTPS proxy server to route TLS outbound requests through.
- no_
proxy str - Comma-separated list of hosts that should not be proxied.
- proxy_
scheme str - The proxy scheme to use when connecting to the proxy server. Default: "http"; must be "http"
- auth Property Map
- Credentials used to authenticate to the proxy server.
- http
Proxy Property Map - HTTP proxy server to route plaintext outbound requests through.
- https
Proxy Property Map - HTTPS proxy server to route TLS outbound requests through.
- no
Proxy String - Comma-separated list of hosts that should not be proxied.
- proxy
Scheme String - The proxy scheme to use when connecting to the proxy server. Default: "http"; must be "http"
AiGatewayModelModelConfigProxyAuth, AiGatewayModelModelConfigProxyAuthArgs
- Password string
- The password to use for proxy authentication. This field is referenceable.
- Username string
- The username to use for proxy authentication. This field is referenceable.
- Password string
- The password to use for proxy authentication. This field is referenceable.
- Username string
- The username to use for proxy authentication. This field is referenceable.
- password string
- The password to use for proxy authentication. This field is referenceable.
- username string
- The username to use for proxy authentication. This field is referenceable.
- password String
- The password to use for proxy authentication. This field is referenceable.
- username String
- The username to use for proxy authentication. This field is referenceable.
- password string
- The password to use for proxy authentication. This field is referenceable.
- username string
- The username to use for proxy authentication. This field is referenceable.
- password str
- The password to use for proxy authentication. This field is referenceable.
- username str
- The username to use for proxy authentication. This field is referenceable.
- password String
- The password to use for proxy authentication. This field is referenceable.
- username String
- The username to use for proxy authentication. This field is referenceable.
AiGatewayModelModelConfigProxyHttpProxy, AiGatewayModelModelConfigProxyHttpProxyArgs
AiGatewayModelModelConfigProxyHttpsProxy, AiGatewayModelModelConfigProxyHttpsProxyArgs
AiGatewayModelModelConfigRoute, AiGatewayModelModelConfigRouteArgs
- Headers Dictionary<string, string>
- One or more lists of values indexed by header name that will cause this route to match if present in the request. The
Hostheader cannot be used with this attribute: hosts should be specified using thehostsattribute. Whenheaderscontains only one value and that value starts with the special prefix~*, the value is interpreted as a regular expression. - Hosts List<string>
- A list of domain names that match this route. Note that the hosts value is case sensitive.
- Https
Redirect doubleStatus Code - The status code Kong responds with when all properties of a route match except the protocol i.e. if the protocol of the request is
HTTPinstead ofHTTPS.Locationheader is injected by Kong if the field is set to 301, 302, 307 or 308. Note: This config applies only if the route is configured to only accept thehttpsprotocol. Default: 426 - Methods List<string>
- A list of HTTP methods that match this route.
- Model
Ai
Gateway Model Model Config Route Model - Configuration for overriding routing to this model using a selector. When no selector location is set, the format default selector is used. When values are not set, the model name is used as the selector value.
- Paths List<string>
- A list of paths that match this route.
- Preserve
Host bool - When matching a route via one of the
hostsdomain names, use the requestHostheader in the upstream request headers. If set tofalse, the upstreamHostheader will be that of the service'shost. Default: false - Protocols List<string>
- An array of the protocols this route should allow. See the route Object section for a list of accepted protocols. When set to only
https, HTTP requests are answered with an upgrade error. When set to onlyhttp, HTTPS requests are answered with an error. Default: ["http","https"] - Regex
Priority double - A number used to choose which route resolves a given request when several routes match it using regexes simultaneously. When two routes match the path and have the same
regex_priority, the older one (lowestcreated_at) is used. Note that the priority for non-regex routes is different (longer non-regex routes are matched before shorter ones). Default: 0 - Request
Buffering bool - Whether to enable request body buffering or not. With HTTP 1.1, it may make sense to turn this off on services that receive data with chunked transfer encoding. Default: true
- Response
Buffering bool - Whether to enable response body buffering or not. With HTTP 1.1, it may make sense to turn this off on services that send data with chunked transfer encoding. Default: true
- Strip
Path bool - When matching a route via one of the
paths, strip the matching prefix from the upstream request URL. Default: true - List<string>
- An optional set of strings associated with the route for grouping and filtering.
- Headers map[string]string
- One or more lists of values indexed by header name that will cause this route to match if present in the request. The
Hostheader cannot be used with this attribute: hosts should be specified using thehostsattribute. Whenheaderscontains only one value and that value starts with the special prefix~*, the value is interpreted as a regular expression. - Hosts []string
- A list of domain names that match this route. Note that the hosts value is case sensitive.
- Https
Redirect float64Status Code - The status code Kong responds with when all properties of a route match except the protocol i.e. if the protocol of the request is
HTTPinstead ofHTTPS.Locationheader is injected by Kong if the field is set to 301, 302, 307 or 308. Note: This config applies only if the route is configured to only accept thehttpsprotocol. Default: 426 - Methods []string
- A list of HTTP methods that match this route.
- Model
Ai
Gateway Model Model Config Route Model - Configuration for overriding routing to this model using a selector. When no selector location is set, the format default selector is used. When values are not set, the model name is used as the selector value.
- Paths []string
- A list of paths that match this route.
- Preserve
Host bool - When matching a route via one of the
hostsdomain names, use the requestHostheader in the upstream request headers. If set tofalse, the upstreamHostheader will be that of the service'shost. Default: false - Protocols []string
- An array of the protocols this route should allow. See the route Object section for a list of accepted protocols. When set to only
https, HTTP requests are answered with an upgrade error. When set to onlyhttp, HTTPS requests are answered with an error. Default: ["http","https"] - Regex
Priority float64 - A number used to choose which route resolves a given request when several routes match it using regexes simultaneously. When two routes match the path and have the same
regex_priority, the older one (lowestcreated_at) is used. Note that the priority for non-regex routes is different (longer non-regex routes are matched before shorter ones). Default: 0 - Request
Buffering bool - Whether to enable request body buffering or not. With HTTP 1.1, it may make sense to turn this off on services that receive data with chunked transfer encoding. Default: true
- Response
Buffering bool - Whether to enable response body buffering or not. With HTTP 1.1, it may make sense to turn this off on services that send data with chunked transfer encoding. Default: true
- Strip
Path bool - When matching a route via one of the
paths, strip the matching prefix from the upstream request URL. Default: true - []string
- An optional set of strings associated with the route for grouping and filtering.
- headers map(string)
- One or more lists of values indexed by header name that will cause this route to match if present in the request. The
Hostheader cannot be used with this attribute: hosts should be specified using thehostsattribute. Whenheaderscontains only one value and that value starts with the special prefix~*, the value is interpreted as a regular expression. - hosts list(string)
- A list of domain names that match this route. Note that the hosts value is case sensitive.
- https_
redirect_ numberstatus_ code - The status code Kong responds with when all properties of a route match except the protocol i.e. if the protocol of the request is
HTTPinstead ofHTTPS.Locationheader is injected by Kong if the field is set to 301, 302, 307 or 308. Note: This config applies only if the route is configured to only accept thehttpsprotocol. Default: 426 - methods list(string)
- A list of HTTP methods that match this route.
- model object
- Configuration for overriding routing to this model using a selector. When no selector location is set, the format default selector is used. When values are not set, the model name is used as the selector value.
- paths list(string)
- A list of paths that match this route.
- preserve_
host bool - When matching a route via one of the
hostsdomain names, use the requestHostheader in the upstream request headers. If set tofalse, the upstreamHostheader will be that of the service'shost. Default: false - protocols list(string)
- An array of the protocols this route should allow. See the route Object section for a list of accepted protocols. When set to only
https, HTTP requests are answered with an upgrade error. When set to onlyhttp, HTTPS requests are answered with an error. Default: ["http","https"] - regex_
priority number - A number used to choose which route resolves a given request when several routes match it using regexes simultaneously. When two routes match the path and have the same
regex_priority, the older one (lowestcreated_at) is used. Note that the priority for non-regex routes is different (longer non-regex routes are matched before shorter ones). Default: 0 - request_
buffering bool - Whether to enable request body buffering or not. With HTTP 1.1, it may make sense to turn this off on services that receive data with chunked transfer encoding. Default: true
- response_
buffering bool - Whether to enable response body buffering or not. With HTTP 1.1, it may make sense to turn this off on services that send data with chunked transfer encoding. Default: true
- strip_
path bool - When matching a route via one of the
paths, strip the matching prefix from the upstream request URL. Default: true - list(string)
- An optional set of strings associated with the route for grouping and filtering.
- headers Map<String,String>
- One or more lists of values indexed by header name that will cause this route to match if present in the request. The
Hostheader cannot be used with this attribute: hosts should be specified using thehostsattribute. Whenheaderscontains only one value and that value starts with the special prefix~*, the value is interpreted as a regular expression. - hosts List<String>
- A list of domain names that match this route. Note that the hosts value is case sensitive.
- https
Redirect DoubleStatus Code - The status code Kong responds with when all properties of a route match except the protocol i.e. if the protocol of the request is
HTTPinstead ofHTTPS.Locationheader is injected by Kong if the field is set to 301, 302, 307 or 308. Note: This config applies only if the route is configured to only accept thehttpsprotocol. Default: 426 - methods List<String>
- A list of HTTP methods that match this route.
- model
Ai
Gateway Model Model Config Route Model - Configuration for overriding routing to this model using a selector. When no selector location is set, the format default selector is used. When values are not set, the model name is used as the selector value.
- paths List<String>
- A list of paths that match this route.
- preserve
Host Boolean - When matching a route via one of the
hostsdomain names, use the requestHostheader in the upstream request headers. If set tofalse, the upstreamHostheader will be that of the service'shost. Default: false - protocols List<String>
- An array of the protocols this route should allow. See the route Object section for a list of accepted protocols. When set to only
https, HTTP requests are answered with an upgrade error. When set to onlyhttp, HTTPS requests are answered with an error. Default: ["http","https"] - regex
Priority Double - A number used to choose which route resolves a given request when several routes match it using regexes simultaneously. When two routes match the path and have the same
regex_priority, the older one (lowestcreated_at) is used. Note that the priority for non-regex routes is different (longer non-regex routes are matched before shorter ones). Default: 0 - request
Buffering Boolean - Whether to enable request body buffering or not. With HTTP 1.1, it may make sense to turn this off on services that receive data with chunked transfer encoding. Default: true
- response
Buffering Boolean - Whether to enable response body buffering or not. With HTTP 1.1, it may make sense to turn this off on services that send data with chunked transfer encoding. Default: true
- strip
Path Boolean - When matching a route via one of the
paths, strip the matching prefix from the upstream request URL. Default: true - List<String>
- An optional set of strings associated with the route for grouping and filtering.
- headers {[key: string]: string}
- One or more lists of values indexed by header name that will cause this route to match if present in the request. The
Hostheader cannot be used with this attribute: hosts should be specified using thehostsattribute. Whenheaderscontains only one value and that value starts with the special prefix~*, the value is interpreted as a regular expression. - hosts string[]
- A list of domain names that match this route. Note that the hosts value is case sensitive.
- https
Redirect numberStatus Code - The status code Kong responds with when all properties of a route match except the protocol i.e. if the protocol of the request is
HTTPinstead ofHTTPS.Locationheader is injected by Kong if the field is set to 301, 302, 307 or 308. Note: This config applies only if the route is configured to only accept thehttpsprotocol. Default: 426 - methods string[]
- A list of HTTP methods that match this route.
- model
Ai
Gateway Model Model Config Route Model - Configuration for overriding routing to this model using a selector. When no selector location is set, the format default selector is used. When values are not set, the model name is used as the selector value.
- paths string[]
- A list of paths that match this route.
- preserve
Host boolean - When matching a route via one of the
hostsdomain names, use the requestHostheader in the upstream request headers. If set tofalse, the upstreamHostheader will be that of the service'shost. Default: false - protocols string[]
- An array of the protocols this route should allow. See the route Object section for a list of accepted protocols. When set to only
https, HTTP requests are answered with an upgrade error. When set to onlyhttp, HTTPS requests are answered with an error. Default: ["http","https"] - regex
Priority number - A number used to choose which route resolves a given request when several routes match it using regexes simultaneously. When two routes match the path and have the same
regex_priority, the older one (lowestcreated_at) is used. Note that the priority for non-regex routes is different (longer non-regex routes are matched before shorter ones). Default: 0 - request
Buffering boolean - Whether to enable request body buffering or not. With HTTP 1.1, it may make sense to turn this off on services that receive data with chunked transfer encoding. Default: true
- response
Buffering boolean - Whether to enable response body buffering or not. With HTTP 1.1, it may make sense to turn this off on services that send data with chunked transfer encoding. Default: true
- strip
Path boolean - When matching a route via one of the
paths, strip the matching prefix from the upstream request URL. Default: true - string[]
- An optional set of strings associated with the route for grouping and filtering.
- headers Mapping[str, str]
- One or more lists of values indexed by header name that will cause this route to match if present in the request. The
Hostheader cannot be used with this attribute: hosts should be specified using thehostsattribute. Whenheaderscontains only one value and that value starts with the special prefix~*, the value is interpreted as a regular expression. - hosts Sequence[str]
- A list of domain names that match this route. Note that the hosts value is case sensitive.
- https_
redirect_ floatstatus_ code - The status code Kong responds with when all properties of a route match except the protocol i.e. if the protocol of the request is
HTTPinstead ofHTTPS.Locationheader is injected by Kong if the field is set to 301, 302, 307 or 308. Note: This config applies only if the route is configured to only accept thehttpsprotocol. Default: 426 - methods Sequence[str]
- A list of HTTP methods that match this route.
- model
Ai
Gateway Model Model Config Route Model - Configuration for overriding routing to this model using a selector. When no selector location is set, the format default selector is used. When values are not set, the model name is used as the selector value.
- paths Sequence[str]
- A list of paths that match this route.
- preserve_
host bool - When matching a route via one of the
hostsdomain names, use the requestHostheader in the upstream request headers. If set tofalse, the upstreamHostheader will be that of the service'shost. Default: false - protocols Sequence[str]
- An array of the protocols this route should allow. See the route Object section for a list of accepted protocols. When set to only
https, HTTP requests are answered with an upgrade error. When set to onlyhttp, HTTPS requests are answered with an error. Default: ["http","https"] - regex_
priority float - A number used to choose which route resolves a given request when several routes match it using regexes simultaneously. When two routes match the path and have the same
regex_priority, the older one (lowestcreated_at) is used. Note that the priority for non-regex routes is different (longer non-regex routes are matched before shorter ones). Default: 0 - request_
buffering bool - Whether to enable request body buffering or not. With HTTP 1.1, it may make sense to turn this off on services that receive data with chunked transfer encoding. Default: true
- response_
buffering bool - Whether to enable response body buffering or not. With HTTP 1.1, it may make sense to turn this off on services that send data with chunked transfer encoding. Default: true
- strip_
path bool - When matching a route via one of the
paths, strip the matching prefix from the upstream request URL. Default: true - Sequence[str]
- An optional set of strings associated with the route for grouping and filtering.
- headers Map<String>
- One or more lists of values indexed by header name that will cause this route to match if present in the request. The
Hostheader cannot be used with this attribute: hosts should be specified using thehostsattribute. Whenheaderscontains only one value and that value starts with the special prefix~*, the value is interpreted as a regular expression. - hosts List<String>
- A list of domain names that match this route. Note that the hosts value is case sensitive.
- https
Redirect NumberStatus Code - The status code Kong responds with when all properties of a route match except the protocol i.e. if the protocol of the request is
HTTPinstead ofHTTPS.Locationheader is injected by Kong if the field is set to 301, 302, 307 or 308. Note: This config applies only if the route is configured to only accept thehttpsprotocol. Default: 426 - methods List<String>
- A list of HTTP methods that match this route.
- model Property Map
- Configuration for overriding routing to this model using a selector. When no selector location is set, the format default selector is used. When values are not set, the model name is used as the selector value.
- paths List<String>
- A list of paths that match this route.
- preserve
Host Boolean - When matching a route via one of the
hostsdomain names, use the requestHostheader in the upstream request headers. If set tofalse, the upstreamHostheader will be that of the service'shost. Default: false - protocols List<String>
- An array of the protocols this route should allow. See the route Object section for a list of accepted protocols. When set to only
https, HTTP requests are answered with an upgrade error. When set to onlyhttp, HTTPS requests are answered with an error. Default: ["http","https"] - regex
Priority Number - A number used to choose which route resolves a given request when several routes match it using regexes simultaneously. When two routes match the path and have the same
regex_priority, the older one (lowestcreated_at) is used. Note that the priority for non-regex routes is different (longer non-regex routes are matched before shorter ones). Default: 0 - request
Buffering Boolean - Whether to enable request body buffering or not. With HTTP 1.1, it may make sense to turn this off on services that receive data with chunked transfer encoding. Default: true
- response
Buffering Boolean - Whether to enable response body buffering or not. With HTTP 1.1, it may make sense to turn this off on services that send data with chunked transfer encoding. Default: true
- strip
Path Boolean - When matching a route via one of the
paths, strip the matching prefix from the upstream request URL. Default: true - List<String>
- An optional set of strings associated with the route for grouping and filtering.
AiGatewayModelModelConfigRouteModel, AiGatewayModelModelConfigRouteModelArgs
- Body
Param string - The body property name to match for routing.
- Header
Param string - The header property name to match for routing.
- Path
Param string - The name of the regex capture group defined in the route path for routing.
- Values List<string>
- An optional model alias. When omitted, the model name is used. When no selector location is configured, the format default selector is used.
- Body
Param string - The body property name to match for routing.
- Header
Param string - The header property name to match for routing.
- Path
Param string - The name of the regex capture group defined in the route path for routing.
- Values []string
- An optional model alias. When omitted, the model name is used. When no selector location is configured, the format default selector is used.
- body_
param string - The body property name to match for routing.
- header_
param string - The header property name to match for routing.
- path_
param string - The name of the regex capture group defined in the route path for routing.
- values list(string)
- An optional model alias. When omitted, the model name is used. When no selector location is configured, the format default selector is used.
- body
Param String - The body property name to match for routing.
- header
Param String - The header property name to match for routing.
- path
Param String - The name of the regex capture group defined in the route path for routing.
- values List<String>
- An optional model alias. When omitted, the model name is used. When no selector location is configured, the format default selector is used.
- body
Param string - The body property name to match for routing.
- header
Param string - The header property name to match for routing.
- path
Param string - The name of the regex capture group defined in the route path for routing.
- values string[]
- An optional model alias. When omitted, the model name is used. When no selector location is configured, the format default selector is used.
- body_
param str - The body property name to match for routing.
- header_
param str - The header property name to match for routing.
- path_
param str - The name of the regex capture group defined in the route path for routing.
- values Sequence[str]
- An optional model alias. When omitted, the model name is used. When no selector location is configured, the format default selector is used.
- body
Param String - The body property name to match for routing.
- header
Param String - The header property name to match for routing.
- path
Param String - The name of the regex capture group defined in the route path for routing.
- values List<String>
- An optional model alias. When omitted, the model name is used. When no selector location is configured, the format default selector is used.
AiGatewayModelModelFormat, AiGatewayModelModelFormatArgs
- Type string
- The format type. possible known values include one of ["anthropic", "bedrock", "cohere", "gemini", "huggingface", "openai"]
- Type string
- The format type. possible known values include one of ["anthropic", "bedrock", "cohere", "gemini", "huggingface", "openai"]
- type string
- The format type. possible known values include one of ["anthropic", "bedrock", "cohere", "gemini", "huggingface", "openai"]
- type String
- The format type. possible known values include one of ["anthropic", "bedrock", "cohere", "gemini", "huggingface", "openai"]
- type string
- The format type. possible known values include one of ["anthropic", "bedrock", "cohere", "gemini", "huggingface", "openai"]
- type str
- The format type. possible known values include one of ["anthropic", "bedrock", "cohere", "gemini", "huggingface", "openai"]
- type String
- The format type. possible known values include one of ["anthropic", "bedrock", "cohere", "gemini", "huggingface", "openai"]
AiGatewayModelModelTarget, AiGatewayModelModelTargetArgs
- Allow
Auth boolOverride - When enabled, request-level auth parameters (such as API keys or bearer tokens) will override the static values defined for the provider. Default: false
- Config
Ai
Gateway Model Model Target Config - Configuration for a target model. Not Null
- Name string
- The name of the model defined in the upstream provider that will be executed. Not Null
- Provider string
- Reference to a model provider instance by name. Not Null
- Semantic
Description string - The semantic description of the target, required if using semantic load balancing. Specially, setting this to 'CATCHALL' will indicate such target to be used when no other targets match the semantic threshold.
- Weight double
- The weight this target gets within the upstream load balancer. Default: 100
- Allow
Auth boolOverride - When enabled, request-level auth parameters (such as API keys or bearer tokens) will override the static values defined for the provider. Default: false
- Config
Ai
Gateway Model Model Target Config - Configuration for a target model. Not Null
- Name string
- The name of the model defined in the upstream provider that will be executed. Not Null
- Provider string
- Reference to a model provider instance by name. Not Null
- Semantic
Description string - The semantic description of the target, required if using semantic load balancing. Specially, setting this to 'CATCHALL' will indicate such target to be used when no other targets match the semantic threshold.
- Weight float64
- The weight this target gets within the upstream load balancer. Default: 100
- allow_
auth_ booloverride - When enabled, request-level auth parameters (such as API keys or bearer tokens) will override the static values defined for the provider. Default: false
- config object
- Configuration for a target model. Not Null
- name string
- The name of the model defined in the upstream provider that will be executed. Not Null
- provider string
- Reference to a model provider instance by name. Not Null
- semantic_
description string - The semantic description of the target, required if using semantic load balancing. Specially, setting this to 'CATCHALL' will indicate such target to be used when no other targets match the semantic threshold.
- weight number
- The weight this target gets within the upstream load balancer. Default: 100
- allow
Auth BooleanOverride - When enabled, request-level auth parameters (such as API keys or bearer tokens) will override the static values defined for the provider. Default: false
- config
Ai
Gateway Model Model Target Config - Configuration for a target model. Not Null
- name String
- The name of the model defined in the upstream provider that will be executed. Not Null
- provider String
- Reference to a model provider instance by name. Not Null
- semantic
Description String - The semantic description of the target, required if using semantic load balancing. Specially, setting this to 'CATCHALL' will indicate such target to be used when no other targets match the semantic threshold.
- weight Double
- The weight this target gets within the upstream load balancer. Default: 100
- allow
Auth booleanOverride - When enabled, request-level auth parameters (such as API keys or bearer tokens) will override the static values defined for the provider. Default: false
- config
Ai
Gateway Model Model Target Config - Configuration for a target model. Not Null
- name string
- The name of the model defined in the upstream provider that will be executed. Not Null
- provider string
- Reference to a model provider instance by name. Not Null
- semantic
Description string - The semantic description of the target, required if using semantic load balancing. Specially, setting this to 'CATCHALL' will indicate such target to be used when no other targets match the semantic threshold.
- weight number
- The weight this target gets within the upstream load balancer. Default: 100
- allow_
auth_ booloverride - When enabled, request-level auth parameters (such as API keys or bearer tokens) will override the static values defined for the provider. Default: false
- config
Ai
Gateway Model Model Target Config - Configuration for a target model. Not Null
- name str
- The name of the model defined in the upstream provider that will be executed. Not Null
- provider str
- Reference to a model provider instance by name. Not Null
- semantic_
description str - The semantic description of the target, required if using semantic load balancing. Specially, setting this to 'CATCHALL' will indicate such target to be used when no other targets match the semantic threshold.
- weight float
- The weight this target gets within the upstream load balancer. Default: 100
- allow
Auth BooleanOverride - When enabled, request-level auth parameters (such as API keys or bearer tokens) will override the static values defined for the provider. Default: false
- config Property Map
- Configuration for a target model. Not Null
- name String
- The name of the model defined in the upstream provider that will be executed. Not Null
- provider String
- Reference to a model provider instance by name. Not Null
- semantic
Description String - The semantic description of the target, required if using semantic load balancing. Specially, setting this to 'CATCHALL' will indicate such target to be used when no other targets match the semantic threshold.
- weight Number
- The weight this target gets within the upstream load balancer. Default: 100
AiGatewayModelModelTargetConfig, AiGatewayModelModelTargetConfigArgs
- Anthropic
Ai
Gateway Model Model Target Config Anthropic - Anthropic-specific configuration for a model.
- Azure
Ai
Gateway Model Model Target Config Azure - Azure-specific configuration for a model.
- Bedrock
Ai
Gateway Model Model Target Config Bedrock - AWS Bedrock-specific configuration for a model.
- Cerebras
Ai
Gateway Model Model Target Config Cerebras - Cerebras-specific configuration for a model.
- Cohere
Ai
Gateway Model Model Target Config Cohere - Cohere-specific configuration for a model.
- Dashscope
Ai
Gateway Model Model Target Config Dashscope - Alibaba DashScope-specific configuration for a model.
- Databricks
Ai
Gateway Model Model Target Config Databricks - Databricks-specific configuration for a model.
- Deepseek
Ai
Gateway Model Model Target Config Deepseek - Deepseek-specific configuration for a model.
- Gemini
Ai
Gateway Model Model Target Config Gemini - Google Gemini-specific configuration for a model.
- Huggingface
Ai
Gateway Model Model Target Config Huggingface - Hugging Face-specific configuration for a model.
- Kimi
Ai
Gateway Model Model Target Config Kimi - Kimi (Moonshot AI)-specific configuration for a model.
- Llama2
Ai
Gateway Model Model Target Config Llama2 - Llama2-specific configuration for a model.
- Mistral
Ai
Gateway Model Model Target Config Mistral - Mistral-specific configuration for a model.
- Ollama
Ai
Gateway Model Model Target Config Ollama - Ollama-specific configuration for a model.
- Openai
Ai
Gateway Model Model Target Config Openai - Openai-specific configuration for a model.
- Sagemaker
Ai
Gateway Model Model Target Config Sagemaker - AWS SageMaker-specific configuration for a model.
- Vercel
Ai
Gateway Model Model Target Config Vercel - Vercel AI Gateway-specific configuration for a model.
- Vllm
Ai
Gateway Model Model Target Config Vllm - Vllm-specific configuration for a model.
- Xai
Ai
Gateway Model Model Target Config Xai - Xai-specific configuration for a model.
- Anthropic
Ai
Gateway Model Model Target Config Anthropic - Anthropic-specific configuration for a model.
- Azure
Ai
Gateway Model Model Target Config Azure - Azure-specific configuration for a model.
- Bedrock
Ai
Gateway Model Model Target Config Bedrock - AWS Bedrock-specific configuration for a model.
- Cerebras
Ai
Gateway Model Model Target Config Cerebras - Cerebras-specific configuration for a model.
- Cohere
Ai
Gateway Model Model Target Config Cohere - Cohere-specific configuration for a model.
- Dashscope
Ai
Gateway Model Model Target Config Dashscope - Alibaba DashScope-specific configuration for a model.
- Databricks
Ai
Gateway Model Model Target Config Databricks - Databricks-specific configuration for a model.
- Deepseek
Ai
Gateway Model Model Target Config Deepseek - Deepseek-specific configuration for a model.
- Gemini
Ai
Gateway Model Model Target Config Gemini - Google Gemini-specific configuration for a model.
- Huggingface
Ai
Gateway Model Model Target Config Huggingface - Hugging Face-specific configuration for a model.
- Kimi
Ai
Gateway Model Model Target Config Kimi - Kimi (Moonshot AI)-specific configuration for a model.
- Llama2
Ai
Gateway Model Model Target Config Llama2 - Llama2-specific configuration for a model.
- Mistral
Ai
Gateway Model Model Target Config Mistral - Mistral-specific configuration for a model.
- Ollama
Ai
Gateway Model Model Target Config Ollama - Ollama-specific configuration for a model.
- Openai
Ai
Gateway Model Model Target Config Openai - Openai-specific configuration for a model.
- Sagemaker
Ai
Gateway Model Model Target Config Sagemaker - AWS SageMaker-specific configuration for a model.
- Vercel
Ai
Gateway Model Model Target Config Vercel - Vercel AI Gateway-specific configuration for a model.
- Vllm
Ai
Gateway Model Model Target Config Vllm - Vllm-specific configuration for a model.
- Xai
Ai
Gateway Model Model Target Config Xai - Xai-specific configuration for a model.
- anthropic object
- Anthropic-specific configuration for a model.
- azure object
- Azure-specific configuration for a model.
- bedrock object
- AWS Bedrock-specific configuration for a model.
- cerebras object
- Cerebras-specific configuration for a model.
- cohere object
- Cohere-specific configuration for a model.
- dashscope object
- Alibaba DashScope-specific configuration for a model.
- databricks object
- Databricks-specific configuration for a model.
- deepseek object
- Deepseek-specific configuration for a model.
- gemini object
- Google Gemini-specific configuration for a model.
- huggingface object
- Hugging Face-specific configuration for a model.
- kimi object
- Kimi (Moonshot AI)-specific configuration for a model.
- llama2 object
- Llama2-specific configuration for a model.
- mistral object
- Mistral-specific configuration for a model.
- ollama object
- Ollama-specific configuration for a model.
- openai object
- Openai-specific configuration for a model.
- sagemaker object
- AWS SageMaker-specific configuration for a model.
- vercel object
- Vercel AI Gateway-specific configuration for a model.
- vllm object
- Vllm-specific configuration for a model.
- xai object
- Xai-specific configuration for a model.
- anthropic
Ai
Gateway Model Model Target Config Anthropic - Anthropic-specific configuration for a model.
- azure
Ai
Gateway Model Model Target Config Azure - Azure-specific configuration for a model.
- bedrock
Ai
Gateway Model Model Target Config Bedrock - AWS Bedrock-specific configuration for a model.
- cerebras
Ai
Gateway Model Model Target Config Cerebras - Cerebras-specific configuration for a model.
- cohere
Ai
Gateway Model Model Target Config Cohere - Cohere-specific configuration for a model.
- dashscope
Ai
Gateway Model Model Target Config Dashscope - Alibaba DashScope-specific configuration for a model.
- databricks
Ai
Gateway Model Model Target Config Databricks - Databricks-specific configuration for a model.
- deepseek
Ai
Gateway Model Model Target Config Deepseek - Deepseek-specific configuration for a model.
- gemini
Ai
Gateway Model Model Target Config Gemini - Google Gemini-specific configuration for a model.
- huggingface
Ai
Gateway Model Model Target Config Huggingface - Hugging Face-specific configuration for a model.
- kimi
Ai
Gateway Model Model Target Config Kimi - Kimi (Moonshot AI)-specific configuration for a model.
- llama2
Ai
Gateway Model Model Target Config Llama2 - Llama2-specific configuration for a model.
- mistral
Ai
Gateway Model Model Target Config Mistral - Mistral-specific configuration for a model.
- ollama
Ai
Gateway Model Model Target Config Ollama - Ollama-specific configuration for a model.
- openai
Ai
Gateway Model Model Target Config Openai - Openai-specific configuration for a model.
- sagemaker
Ai
Gateway Model Model Target Config Sagemaker - AWS SageMaker-specific configuration for a model.
- vercel
Ai
Gateway Model Model Target Config Vercel - Vercel AI Gateway-specific configuration for a model.
- vllm
Ai
Gateway Model Model Target Config Vllm - Vllm-specific configuration for a model.
- xai
Ai
Gateway Model Model Target Config Xai - Xai-specific configuration for a model.
- anthropic
Ai
Gateway Model Model Target Config Anthropic - Anthropic-specific configuration for a model.
- azure
Ai
Gateway Model Model Target Config Azure - Azure-specific configuration for a model.
- bedrock
Ai
Gateway Model Model Target Config Bedrock - AWS Bedrock-specific configuration for a model.
- cerebras
Ai
Gateway Model Model Target Config Cerebras - Cerebras-specific configuration for a model.
- cohere
Ai
Gateway Model Model Target Config Cohere - Cohere-specific configuration for a model.
- dashscope
Ai
Gateway Model Model Target Config Dashscope - Alibaba DashScope-specific configuration for a model.
- databricks
Ai
Gateway Model Model Target Config Databricks - Databricks-specific configuration for a model.
- deepseek
Ai
Gateway Model Model Target Config Deepseek - Deepseek-specific configuration for a model.
- gemini
Ai
Gateway Model Model Target Config Gemini - Google Gemini-specific configuration for a model.
- huggingface
Ai
Gateway Model Model Target Config Huggingface - Hugging Face-specific configuration for a model.
- kimi
Ai
Gateway Model Model Target Config Kimi - Kimi (Moonshot AI)-specific configuration for a model.
- llama2
Ai
Gateway Model Model Target Config Llama2 - Llama2-specific configuration for a model.
- mistral
Ai
Gateway Model Model Target Config Mistral - Mistral-specific configuration for a model.
- ollama
Ai
Gateway Model Model Target Config Ollama - Ollama-specific configuration for a model.
- openai
Ai
Gateway Model Model Target Config Openai - Openai-specific configuration for a model.
- sagemaker
Ai
Gateway Model Model Target Config Sagemaker - AWS SageMaker-specific configuration for a model.
- vercel
Ai
Gateway Model Model Target Config Vercel - Vercel AI Gateway-specific configuration for a model.
- vllm
Ai
Gateway Model Model Target Config Vllm - Vllm-specific configuration for a model.
- xai
Ai
Gateway Model Model Target Config Xai - Xai-specific configuration for a model.
- anthropic
Ai
Gateway Model Model Target Config Anthropic - Anthropic-specific configuration for a model.
- azure
Ai
Gateway Model Model Target Config Azure - Azure-specific configuration for a model.
- bedrock
Ai
Gateway Model Model Target Config Bedrock - AWS Bedrock-specific configuration for a model.
- cerebras
Ai
Gateway Model Model Target Config Cerebras - Cerebras-specific configuration for a model.
- cohere
Ai
Gateway Model Model Target Config Cohere - Cohere-specific configuration for a model.
- dashscope
Ai
Gateway Model Model Target Config Dashscope - Alibaba DashScope-specific configuration for a model.
- databricks
Ai
Gateway Model Model Target Config Databricks - Databricks-specific configuration for a model.
- deepseek
Ai
Gateway Model Model Target Config Deepseek - Deepseek-specific configuration for a model.
- gemini
Ai
Gateway Model Model Target Config Gemini - Google Gemini-specific configuration for a model.
- huggingface
Ai
Gateway Model Model Target Config Huggingface - Hugging Face-specific configuration for a model.
- kimi
Ai
Gateway Model Model Target Config Kimi - Kimi (Moonshot AI)-specific configuration for a model.
- llama2
Ai
Gateway Model Model Target Config Llama2 - Llama2-specific configuration for a model.
- mistral
Ai
Gateway Model Model Target Config Mistral - Mistral-specific configuration for a model.
- ollama
Ai
Gateway Model Model Target Config Ollama - Ollama-specific configuration for a model.
- openai
Ai
Gateway Model Model Target Config Openai - Openai-specific configuration for a model.
- sagemaker
Ai
Gateway Model Model Target Config Sagemaker - AWS SageMaker-specific configuration for a model.
- vercel
Ai
Gateway Model Model Target Config Vercel - Vercel AI Gateway-specific configuration for a model.
- vllm
Ai
Gateway Model Model Target Config Vllm - Vllm-specific configuration for a model.
- xai
Ai
Gateway Model Model Target Config Xai - Xai-specific configuration for a model.
- anthropic Property Map
- Anthropic-specific configuration for a model.
- azure Property Map
- Azure-specific configuration for a model.
- bedrock Property Map
- AWS Bedrock-specific configuration for a model.
- cerebras Property Map
- Cerebras-specific configuration for a model.
- cohere Property Map
- Cohere-specific configuration for a model.
- dashscope Property Map
- Alibaba DashScope-specific configuration for a model.
- databricks Property Map
- Databricks-specific configuration for a model.
- deepseek Property Map
- Deepseek-specific configuration for a model.
- gemini Property Map
- Google Gemini-specific configuration for a model.
- huggingface Property Map
- Hugging Face-specific configuration for a model.
- kimi Property Map
- Kimi (Moonshot AI)-specific configuration for a model.
- llama2 Property Map
- Llama2-specific configuration for a model.
- mistral Property Map
- Mistral-specific configuration for a model.
- ollama Property Map
- Ollama-specific configuration for a model.
- openai Property Map
- Openai-specific configuration for a model.
- sagemaker Property Map
- AWS SageMaker-specific configuration for a model.
- vercel Property Map
- Vercel AI Gateway-specific configuration for a model.
- vllm Property Map
- Vllm-specific configuration for a model.
- xai Property Map
- Xai-specific configuration for a model.
AiGatewayModelModelTargetConfigAnthropic, AiGatewayModelModelTargetConfigAnthropicArgs
- Cache
Read doubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write doubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write List<AiCost Lists Gateway Model Model Target Config Anthropic Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window List<AiFactors Gateway Model Model Target Config Anthropic Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions double - The number of dimensions for embedding outputs.
- Input
Cost double - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens double - The maximum number of tokens to generate in the response.
- Output
Cost double - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier List<AiFactors Gateway Model Model Target Config Anthropic Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature double
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K double - Limits the number of highest-probability tokens considered during generation.
- Top
P double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- Version string
- The Anthropic API version to use. Default: "2023-06-01"
- Cache
Read float64Cost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write float64Cost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write []AiCost Lists Gateway Model Model Target Config Anthropic Cache Write Cost List - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window []AiFactors Gateway Model Model Target Config Anthropic Context Window Factor - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions float64 - The number of dimensions for embedding outputs.
- Input
Cost float64 - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens float64 - The maximum number of tokens to generate in the response.
- Output
Cost float64 - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier []AiFactors Gateway Model Model Target Config Anthropic Service Tier Factor - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature float64
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K float64 - Limits the number of highest-probability tokens considered during generation.
- Top
P float64 - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- Version string
- The Anthropic API version to use. Default: "2023-06-01"
- cache_
read_ numbercost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ numbercost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ list(object)cost_ lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ list(object)factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions number - The number of dimensions for embedding outputs.
- input_
cost number - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens number - The maximum number of tokens to generate in the response.
- output_
cost number - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ list(object)factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k number - Limits the number of highest-probability tokens considered during generation.
- top_
p number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url string - The upstream URL for the model endpoint.
- version string
- The Anthropic API version to use. Default: "2023-06-01"
- cache
Read DoubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write DoubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<AiCost Lists Gateway Model Model Target Config Anthropic Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<AiFactors Gateway Model Model Target Config Anthropic Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Double - The number of dimensions for embedding outputs.
- input
Cost Double - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Double - The maximum number of tokens to generate in the response.
- output
Cost Double - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<AiFactors Gateway Model Model Target Config Anthropic Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Double
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Double - Limits the number of highest-probability tokens considered during generation.
- top
P Double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
- version String
- The Anthropic API version to use. Default: "2023-06-01"
- cache
Read numberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write numberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write AiCost Lists Gateway Model Model Target Config Anthropic Cache Write Cost List[] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window AiFactors Gateway Model Model Target Config Anthropic Context Window Factor[] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions number - The number of dimensions for embedding outputs.
- input
Cost number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens number - The maximum number of tokens to generate in the response.
- output
Cost number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier AiFactors Gateway Model Model Target Config Anthropic Service Tier Factor[] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K number - Limits the number of highest-probability tokens considered during generation.
- top
P number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url string - The upstream URL for the model endpoint.
- version string
- The Anthropic API version to use. Default: "2023-06-01"
- cache_
read_ floatcost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ floatcost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ Sequence[Aicost_ lists Gateway Model Model Target Config Anthropic Cache Write Cost List] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ Sequence[Aifactors Gateway Model Model Target Config Anthropic Context Window Factor] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions float - The number of dimensions for embedding outputs.
- input_
cost float - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens float - The maximum number of tokens to generate in the response.
- output_
cost float - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ Sequence[Aifactors Gateway Model Model Target Config Anthropic Service Tier Factor] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature float
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k float - Limits the number of highest-probability tokens considered during generation.
- top_
p float - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url str - The upstream URL for the model endpoint.
- version str
- The Anthropic API version to use. Default: "2023-06-01"
- cache
Read NumberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write NumberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<Property Map>Cost Lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<Property Map>Factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Number - The number of dimensions for embedding outputs.
- input
Cost Number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Number - The maximum number of tokens to generate in the response.
- output
Cost Number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<Property Map>Factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Number - Limits the number of highest-probability tokens considered during generation.
- top
P Number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
- version String
- The Anthropic API version to use. Default: "2023-06-01"
AiGatewayModelModelTargetConfigAnthropicCacheWriteCostList, AiGatewayModelModelTargetConfigAnthropicCacheWriteCostListArgs
AiGatewayModelModelTargetConfigAnthropicContextWindowFactor, AiGatewayModelModelTargetConfigAnthropicContextWindowFactorArgs
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor double - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor double - Multiplier applied to output pricing above the threshold. Not Null
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor float64 - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor float64 - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor number - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor number - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Double - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Double - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor number - Multiplier applied to output pricing above the threshold. Not Null
- above str
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor float - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor float - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Number - Multiplier applied to output pricing above the threshold. Not Null
AiGatewayModelModelTargetConfigAnthropicServiceTierFactor, AiGatewayModelModelTargetConfigAnthropicServiceTierFactorArgs
- Factor double
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- Factor float64
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Double
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor float
- Multiplier applied to the whole request for this service tier. Not Null
- tier str
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Number
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
AiGatewayModelModelTargetConfigAzure, AiGatewayModelModelTargetConfigAzureArgs
- Api
Version string - The Azure OpenAI API version to use. Default: "2023-05-15"
- Cache
Read doubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write doubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write List<AiCost Lists Gateway Model Model Target Config Azure Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window List<AiFactors Gateway Model Model Target Config Azure Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Deployment
Id string - The Azure deployment ID for the model. Applies when the Azure provider's
serviceisazure-openai; not used forazure-foundry. - Embeddings
Dimensions double - The number of dimensions for embedding outputs.
- Foundry
Path stringPrefix - The API path prefix for the Azure AI Foundry endpoint, selecting the model's
API surface.
/openai/v1targets the OpenAI-compatible surface;/anthropic/v1targets the Anthropic surface. Applies when the Azure provider'sserviceisazure-foundry. possible known values include one of ["/openai/v1", "/anthropic/v1"]; Default: "/openai/v1" - Input
Cost double - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens double - The maximum number of tokens to generate in the response.
- Output
Cost double - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier List<AiFactors Gateway Model Model Target Config Azure Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature double
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K double - Limits the number of highest-probability tokens considered during generation.
- Top
P double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- Api
Version string - The Azure OpenAI API version to use. Default: "2023-05-15"
- Cache
Read float64Cost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write float64Cost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write []AiCost Lists Gateway Model Model Target Config Azure Cache Write Cost List - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window []AiFactors Gateway Model Model Target Config Azure Context Window Factor - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Deployment
Id string - The Azure deployment ID for the model. Applies when the Azure provider's
serviceisazure-openai; not used forazure-foundry. - Embeddings
Dimensions float64 - The number of dimensions for embedding outputs.
- Foundry
Path stringPrefix - The API path prefix for the Azure AI Foundry endpoint, selecting the model's
API surface.
/openai/v1targets the OpenAI-compatible surface;/anthropic/v1targets the Anthropic surface. Applies when the Azure provider'sserviceisazure-foundry. possible known values include one of ["/openai/v1", "/anthropic/v1"]; Default: "/openai/v1" - Input
Cost float64 - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens float64 - The maximum number of tokens to generate in the response.
- Output
Cost float64 - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier []AiFactors Gateway Model Model Target Config Azure Service Tier Factor - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature float64
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K float64 - Limits the number of highest-probability tokens considered during generation.
- Top
P float64 - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- api_
version string - The Azure OpenAI API version to use. Default: "2023-05-15"
- cache_
read_ numbercost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ numbercost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ list(object)cost_ lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ list(object)factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- deployment_
id string - The Azure deployment ID for the model. Applies when the Azure provider's
serviceisazure-openai; not used forazure-foundry. - embeddings_
dimensions number - The number of dimensions for embedding outputs.
- foundry_
path_ stringprefix - The API path prefix for the Azure AI Foundry endpoint, selecting the model's
API surface.
/openai/v1targets the OpenAI-compatible surface;/anthropic/v1targets the Anthropic surface. Applies when the Azure provider'sserviceisazure-foundry. possible known values include one of ["/openai/v1", "/anthropic/v1"]; Default: "/openai/v1" - input_
cost number - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens number - The maximum number of tokens to generate in the response.
- output_
cost number - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ list(object)factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k number - Limits the number of highest-probability tokens considered during generation.
- top_
p number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url string - The upstream URL for the model endpoint.
- api
Version String - The Azure OpenAI API version to use. Default: "2023-05-15"
- cache
Read DoubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write DoubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<AiCost Lists Gateway Model Model Target Config Azure Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<AiFactors Gateway Model Model Target Config Azure Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- deployment
Id String - The Azure deployment ID for the model. Applies when the Azure provider's
serviceisazure-openai; not used forazure-foundry. - embeddings
Dimensions Double - The number of dimensions for embedding outputs.
- foundry
Path StringPrefix - The API path prefix for the Azure AI Foundry endpoint, selecting the model's
API surface.
/openai/v1targets the OpenAI-compatible surface;/anthropic/v1targets the Anthropic surface. Applies when the Azure provider'sserviceisazure-foundry. possible known values include one of ["/openai/v1", "/anthropic/v1"]; Default: "/openai/v1" - input
Cost Double - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Double - The maximum number of tokens to generate in the response.
- output
Cost Double - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<AiFactors Gateway Model Model Target Config Azure Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Double
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Double - Limits the number of highest-probability tokens considered during generation.
- top
P Double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
- api
Version string - The Azure OpenAI API version to use. Default: "2023-05-15"
- cache
Read numberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write numberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write AiCost Lists Gateway Model Model Target Config Azure Cache Write Cost List[] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window AiFactors Gateway Model Model Target Config Azure Context Window Factor[] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- deployment
Id string - The Azure deployment ID for the model. Applies when the Azure provider's
serviceisazure-openai; not used forazure-foundry. - embeddings
Dimensions number - The number of dimensions for embedding outputs.
- foundry
Path stringPrefix - The API path prefix for the Azure AI Foundry endpoint, selecting the model's
API surface.
/openai/v1targets the OpenAI-compatible surface;/anthropic/v1targets the Anthropic surface. Applies when the Azure provider'sserviceisazure-foundry. possible known values include one of ["/openai/v1", "/anthropic/v1"]; Default: "/openai/v1" - input
Cost number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens number - The maximum number of tokens to generate in the response.
- output
Cost number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier AiFactors Gateway Model Model Target Config Azure Service Tier Factor[] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K number - Limits the number of highest-probability tokens considered during generation.
- top
P number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url string - The upstream URL for the model endpoint.
- api_
version str - The Azure OpenAI API version to use. Default: "2023-05-15"
- cache_
read_ floatcost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ floatcost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ Sequence[Aicost_ lists Gateway Model Model Target Config Azure Cache Write Cost List] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ Sequence[Aifactors Gateway Model Model Target Config Azure Context Window Factor] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- deployment_
id str - The Azure deployment ID for the model. Applies when the Azure provider's
serviceisazure-openai; not used forazure-foundry. - embeddings_
dimensions float - The number of dimensions for embedding outputs.
- foundry_
path_ strprefix - The API path prefix for the Azure AI Foundry endpoint, selecting the model's
API surface.
/openai/v1targets the OpenAI-compatible surface;/anthropic/v1targets the Anthropic surface. Applies when the Azure provider'sserviceisazure-foundry. possible known values include one of ["/openai/v1", "/anthropic/v1"]; Default: "/openai/v1" - input_
cost float - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens float - The maximum number of tokens to generate in the response.
- output_
cost float - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ Sequence[Aifactors Gateway Model Model Target Config Azure Service Tier Factor] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature float
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k float - Limits the number of highest-probability tokens considered during generation.
- top_
p float - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url str - The upstream URL for the model endpoint.
- api
Version String - The Azure OpenAI API version to use. Default: "2023-05-15"
- cache
Read NumberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write NumberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<Property Map>Cost Lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<Property Map>Factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- deployment
Id String - The Azure deployment ID for the model. Applies when the Azure provider's
serviceisazure-openai; not used forazure-foundry. - embeddings
Dimensions Number - The number of dimensions for embedding outputs.
- foundry
Path StringPrefix - The API path prefix for the Azure AI Foundry endpoint, selecting the model's
API surface.
/openai/v1targets the OpenAI-compatible surface;/anthropic/v1targets the Anthropic surface. Applies when the Azure provider'sserviceisazure-foundry. possible known values include one of ["/openai/v1", "/anthropic/v1"]; Default: "/openai/v1" - input
Cost Number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Number - The maximum number of tokens to generate in the response.
- output
Cost Number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<Property Map>Factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Number - Limits the number of highest-probability tokens considered during generation.
- top
P Number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
AiGatewayModelModelTargetConfigAzureCacheWriteCostList, AiGatewayModelModelTargetConfigAzureCacheWriteCostListArgs
AiGatewayModelModelTargetConfigAzureContextWindowFactor, AiGatewayModelModelTargetConfigAzureContextWindowFactorArgs
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor double - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor double - Multiplier applied to output pricing above the threshold. Not Null
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor float64 - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor float64 - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor number - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor number - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Double - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Double - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor number - Multiplier applied to output pricing above the threshold. Not Null
- above str
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor float - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor float - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Number - Multiplier applied to output pricing above the threshold. Not Null
AiGatewayModelModelTargetConfigAzureServiceTierFactor, AiGatewayModelModelTargetConfigAzureServiceTierFactorArgs
- Factor double
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- Factor float64
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Double
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor float
- Multiplier applied to the whole request for this service tier. Not Null
- tier str
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Number
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
AiGatewayModelModelTargetConfigBedrock, AiGatewayModelModelTargetConfigBedrockArgs
- Batch
Bucket stringPrefix - S3 bucket prefix for batch inference jobs.
- Cache
Read doubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write doubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write List<AiCost Lists Gateway Model Model Target Config Bedrock Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window List<AiFactors Gateway Model Model Target Config Bedrock Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions double - The number of dimensions for embedding outputs.
- Embeddings
Normalize bool - Whether to normalize embedding vectors in the response. Default: false
- Input
Cost double - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens double - The maximum number of tokens to generate in the response.
- Output
Cost double - Cost per 1M output tokens for billing and cost tracking.
- Performance
Config stringLatency - Latency performance configuration for the model invocation.
- Region string
- The AWS region for the model. Setting this option overrides the AWS_REGION environment variable.
- Service
Tier List<AiFactors Gateway Model Model Target Config Bedrock Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature double
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K double - Limits the number of highest-probability tokens considered during generation.
- Top
P double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- Video
Output stringS3Uri - S3 URI for storing video generation outputs.
- Batch
Bucket stringPrefix - S3 bucket prefix for batch inference jobs.
- Cache
Read float64Cost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write float64Cost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write []AiCost Lists Gateway Model Model Target Config Bedrock Cache Write Cost List - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window []AiFactors Gateway Model Model Target Config Bedrock Context Window Factor - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions float64 - The number of dimensions for embedding outputs.
- Embeddings
Normalize bool - Whether to normalize embedding vectors in the response. Default: false
- Input
Cost float64 - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens float64 - The maximum number of tokens to generate in the response.
- Output
Cost float64 - Cost per 1M output tokens for billing and cost tracking.
- Performance
Config stringLatency - Latency performance configuration for the model invocation.
- Region string
- The AWS region for the model. Setting this option overrides the AWS_REGION environment variable.
- Service
Tier []AiFactors Gateway Model Model Target Config Bedrock Service Tier Factor - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature float64
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K float64 - Limits the number of highest-probability tokens considered during generation.
- Top
P float64 - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- Video
Output stringS3Uri - S3 URI for storing video generation outputs.
- batch_
bucket_ stringprefix - S3 bucket prefix for batch inference jobs.
- cache_
read_ numbercost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ numbercost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ list(object)cost_ lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ list(object)factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions number - The number of dimensions for embedding outputs.
- embeddings_
normalize bool - Whether to normalize embedding vectors in the response. Default: false
- input_
cost number - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens number - The maximum number of tokens to generate in the response.
- output_
cost number - Cost per 1M output tokens for billing and cost tracking.
- performance_
config_ stringlatency - Latency performance configuration for the model invocation.
- region string
- The AWS region for the model. Setting this option overrides the AWS_REGION environment variable.
- service_
tier_ list(object)factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k number - Limits the number of highest-probability tokens considered during generation.
- top_
p number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url string - The upstream URL for the model endpoint.
- video_
output_ strings3_ uri - S3 URI for storing video generation outputs.
- batch
Bucket StringPrefix - S3 bucket prefix for batch inference jobs.
- cache
Read DoubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write DoubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<AiCost Lists Gateway Model Model Target Config Bedrock Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<AiFactors Gateway Model Model Target Config Bedrock Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Double - The number of dimensions for embedding outputs.
- embeddings
Normalize Boolean - Whether to normalize embedding vectors in the response. Default: false
- input
Cost Double - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Double - The maximum number of tokens to generate in the response.
- output
Cost Double - Cost per 1M output tokens for billing and cost tracking.
- performance
Config StringLatency - Latency performance configuration for the model invocation.
- region String
- The AWS region for the model. Setting this option overrides the AWS_REGION environment variable.
- service
Tier List<AiFactors Gateway Model Model Target Config Bedrock Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Double
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Double - Limits the number of highest-probability tokens considered during generation.
- top
P Double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
- video
Output StringS3Uri - S3 URI for storing video generation outputs.
- batch
Bucket stringPrefix - S3 bucket prefix for batch inference jobs.
- cache
Read numberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write numberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write AiCost Lists Gateway Model Model Target Config Bedrock Cache Write Cost List[] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window AiFactors Gateway Model Model Target Config Bedrock Context Window Factor[] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions number - The number of dimensions for embedding outputs.
- embeddings
Normalize boolean - Whether to normalize embedding vectors in the response. Default: false
- input
Cost number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens number - The maximum number of tokens to generate in the response.
- output
Cost number - Cost per 1M output tokens for billing and cost tracking.
- performance
Config stringLatency - Latency performance configuration for the model invocation.
- region string
- The AWS region for the model. Setting this option overrides the AWS_REGION environment variable.
- service
Tier AiFactors Gateway Model Model Target Config Bedrock Service Tier Factor[] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K number - Limits the number of highest-probability tokens considered during generation.
- top
P number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url string - The upstream URL for the model endpoint.
- video
Output stringS3Uri - S3 URI for storing video generation outputs.
- batch_
bucket_ strprefix - S3 bucket prefix for batch inference jobs.
- cache_
read_ floatcost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ floatcost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ Sequence[Aicost_ lists Gateway Model Model Target Config Bedrock Cache Write Cost List] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ Sequence[Aifactors Gateway Model Model Target Config Bedrock Context Window Factor] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions float - The number of dimensions for embedding outputs.
- embeddings_
normalize bool - Whether to normalize embedding vectors in the response. Default: false
- input_
cost float - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens float - The maximum number of tokens to generate in the response.
- output_
cost float - Cost per 1M output tokens for billing and cost tracking.
- performance_
config_ strlatency - Latency performance configuration for the model invocation.
- region str
- The AWS region for the model. Setting this option overrides the AWS_REGION environment variable.
- service_
tier_ Sequence[Aifactors Gateway Model Model Target Config Bedrock Service Tier Factor] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature float
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k float - Limits the number of highest-probability tokens considered during generation.
- top_
p float - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url str - The upstream URL for the model endpoint.
- video_
output_ strs3_ uri - S3 URI for storing video generation outputs.
- batch
Bucket StringPrefix - S3 bucket prefix for batch inference jobs.
- cache
Read NumberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write NumberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<Property Map>Cost Lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<Property Map>Factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Number - The number of dimensions for embedding outputs.
- embeddings
Normalize Boolean - Whether to normalize embedding vectors in the response. Default: false
- input
Cost Number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Number - The maximum number of tokens to generate in the response.
- output
Cost Number - Cost per 1M output tokens for billing and cost tracking.
- performance
Config StringLatency - Latency performance configuration for the model invocation.
- region String
- The AWS region for the model. Setting this option overrides the AWS_REGION environment variable.
- service
Tier List<Property Map>Factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Number - Limits the number of highest-probability tokens considered during generation.
- top
P Number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
- video
Output StringS3Uri - S3 URI for storing video generation outputs.
AiGatewayModelModelTargetConfigBedrockCacheWriteCostList, AiGatewayModelModelTargetConfigBedrockCacheWriteCostListArgs
AiGatewayModelModelTargetConfigBedrockContextWindowFactor, AiGatewayModelModelTargetConfigBedrockContextWindowFactorArgs
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor double - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor double - Multiplier applied to output pricing above the threshold. Not Null
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor float64 - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor float64 - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor number - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor number - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Double - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Double - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor number - Multiplier applied to output pricing above the threshold. Not Null
- above str
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor float - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor float - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Number - Multiplier applied to output pricing above the threshold. Not Null
AiGatewayModelModelTargetConfigBedrockServiceTierFactor, AiGatewayModelModelTargetConfigBedrockServiceTierFactorArgs
- Factor double
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- Factor float64
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Double
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor float
- Multiplier applied to the whole request for this service tier. Not Null
- tier str
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Number
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
AiGatewayModelModelTargetConfigCerebras, AiGatewayModelModelTargetConfigCerebrasArgs
- Cache
Read doubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write doubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write List<AiCost Lists Gateway Model Model Target Config Cerebras Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window List<AiFactors Gateway Model Model Target Config Cerebras Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions double - The number of dimensions for embedding outputs.
- Input
Cost double - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens double - The maximum number of tokens to generate in the response.
- Output
Cost double - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier List<AiFactors Gateway Model Model Target Config Cerebras Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature double
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K double - Limits the number of highest-probability tokens considered during generation.
- Top
P double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- Cache
Read float64Cost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write float64Cost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write []AiCost Lists Gateway Model Model Target Config Cerebras Cache Write Cost List - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window []AiFactors Gateway Model Model Target Config Cerebras Context Window Factor - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions float64 - The number of dimensions for embedding outputs.
- Input
Cost float64 - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens float64 - The maximum number of tokens to generate in the response.
- Output
Cost float64 - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier []AiFactors Gateway Model Model Target Config Cerebras Service Tier Factor - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature float64
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K float64 - Limits the number of highest-probability tokens considered during generation.
- Top
P float64 - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- cache_
read_ numbercost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ numbercost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ list(object)cost_ lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ list(object)factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions number - The number of dimensions for embedding outputs.
- input_
cost number - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens number - The maximum number of tokens to generate in the response.
- output_
cost number - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ list(object)factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k number - Limits the number of highest-probability tokens considered during generation.
- top_
p number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url string - The upstream URL for the model endpoint.
- cache
Read DoubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write DoubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<AiCost Lists Gateway Model Model Target Config Cerebras Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<AiFactors Gateway Model Model Target Config Cerebras Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Double - The number of dimensions for embedding outputs.
- input
Cost Double - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Double - The maximum number of tokens to generate in the response.
- output
Cost Double - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<AiFactors Gateway Model Model Target Config Cerebras Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Double
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Double - Limits the number of highest-probability tokens considered during generation.
- top
P Double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
- cache
Read numberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write numberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write AiCost Lists Gateway Model Model Target Config Cerebras Cache Write Cost List[] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window AiFactors Gateway Model Model Target Config Cerebras Context Window Factor[] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions number - The number of dimensions for embedding outputs.
- input
Cost number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens number - The maximum number of tokens to generate in the response.
- output
Cost number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier AiFactors Gateway Model Model Target Config Cerebras Service Tier Factor[] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K number - Limits the number of highest-probability tokens considered during generation.
- top
P number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url string - The upstream URL for the model endpoint.
- cache_
read_ floatcost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ floatcost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ Sequence[Aicost_ lists Gateway Model Model Target Config Cerebras Cache Write Cost List] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ Sequence[Aifactors Gateway Model Model Target Config Cerebras Context Window Factor] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions float - The number of dimensions for embedding outputs.
- input_
cost float - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens float - The maximum number of tokens to generate in the response.
- output_
cost float - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ Sequence[Aifactors Gateway Model Model Target Config Cerebras Service Tier Factor] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature float
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k float - Limits the number of highest-probability tokens considered during generation.
- top_
p float - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url str - The upstream URL for the model endpoint.
- cache
Read NumberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write NumberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<Property Map>Cost Lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<Property Map>Factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Number - The number of dimensions for embedding outputs.
- input
Cost Number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Number - The maximum number of tokens to generate in the response.
- output
Cost Number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<Property Map>Factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Number - Limits the number of highest-probability tokens considered during generation.
- top
P Number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
AiGatewayModelModelTargetConfigCerebrasCacheWriteCostList, AiGatewayModelModelTargetConfigCerebrasCacheWriteCostListArgs
AiGatewayModelModelTargetConfigCerebrasContextWindowFactor, AiGatewayModelModelTargetConfigCerebrasContextWindowFactorArgs
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor double - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor double - Multiplier applied to output pricing above the threshold. Not Null
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor float64 - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor float64 - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor number - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor number - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Double - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Double - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor number - Multiplier applied to output pricing above the threshold. Not Null
- above str
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor float - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor float - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Number - Multiplier applied to output pricing above the threshold. Not Null
AiGatewayModelModelTargetConfigCerebrasServiceTierFactor, AiGatewayModelModelTargetConfigCerebrasServiceTierFactorArgs
- Factor double
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- Factor float64
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Double
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor float
- Multiplier applied to the whole request for this service tier. Not Null
- tier str
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Number
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
AiGatewayModelModelTargetConfigCohere, AiGatewayModelModelTargetConfigCohereArgs
- Api
Version string - Cohere API version.
v1uses the legacy/v1/chatendpoint;v2(default) uses/v2/chatand supports tool calling. possible known values include one of ["v1", "v2"]; Default: "v2" - Cache
Read doubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write doubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write List<AiCost Lists Gateway Model Model Target Config Cohere Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window List<AiFactors Gateway Model Model Target Config Cohere Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embedding
Input stringType - The intended downstream use of the embeddings to improve model quality. possible known values include one of ["classification", "clustering", "image", "searchdocument", "searchquery"]; Default: "classification"
- Embeddings
Dimensions double - The number of dimensions for embedding outputs.
- Input
Cost double - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens double - The maximum number of tokens to generate in the response.
- Output
Cost double - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier List<AiFactors Gateway Model Model Target Config Cohere Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature double
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K double - Limits the number of highest-probability tokens considered during generation.
- Top
P double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- Wait
For boolModel - Whether to wait for the model to be ready before sending the request. Default: false
- Api
Version string - Cohere API version.
v1uses the legacy/v1/chatendpoint;v2(default) uses/v2/chatand supports tool calling. possible known values include one of ["v1", "v2"]; Default: "v2" - Cache
Read float64Cost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write float64Cost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write []AiCost Lists Gateway Model Model Target Config Cohere Cache Write Cost List - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window []AiFactors Gateway Model Model Target Config Cohere Context Window Factor - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embedding
Input stringType - The intended downstream use of the embeddings to improve model quality. possible known values include one of ["classification", "clustering", "image", "searchdocument", "searchquery"]; Default: "classification"
- Embeddings
Dimensions float64 - The number of dimensions for embedding outputs.
- Input
Cost float64 - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens float64 - The maximum number of tokens to generate in the response.
- Output
Cost float64 - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier []AiFactors Gateway Model Model Target Config Cohere Service Tier Factor - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature float64
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K float64 - Limits the number of highest-probability tokens considered during generation.
- Top
P float64 - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- Wait
For boolModel - Whether to wait for the model to be ready before sending the request. Default: false
- api_
version string - Cohere API version.
v1uses the legacy/v1/chatendpoint;v2(default) uses/v2/chatand supports tool calling. possible known values include one of ["v1", "v2"]; Default: "v2" - cache_
read_ numbercost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ numbercost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ list(object)cost_ lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ list(object)factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embedding_
input_ stringtype - The intended downstream use of the embeddings to improve model quality. possible known values include one of ["classification", "clustering", "image", "searchdocument", "searchquery"]; Default: "classification"
- embeddings_
dimensions number - The number of dimensions for embedding outputs.
- input_
cost number - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens number - The maximum number of tokens to generate in the response.
- output_
cost number - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ list(object)factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k number - Limits the number of highest-probability tokens considered during generation.
- top_
p number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url string - The upstream URL for the model endpoint.
- wait_
for_ boolmodel - Whether to wait for the model to be ready before sending the request. Default: false
- api
Version String - Cohere API version.
v1uses the legacy/v1/chatendpoint;v2(default) uses/v2/chatand supports tool calling. possible known values include one of ["v1", "v2"]; Default: "v2" - cache
Read DoubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write DoubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<AiCost Lists Gateway Model Model Target Config Cohere Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<AiFactors Gateway Model Model Target Config Cohere Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embedding
Input StringType - The intended downstream use of the embeddings to improve model quality. possible known values include one of ["classification", "clustering", "image", "searchdocument", "searchquery"]; Default: "classification"
- embeddings
Dimensions Double - The number of dimensions for embedding outputs.
- input
Cost Double - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Double - The maximum number of tokens to generate in the response.
- output
Cost Double - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<AiFactors Gateway Model Model Target Config Cohere Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Double
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Double - Limits the number of highest-probability tokens considered during generation.
- top
P Double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
- wait
For BooleanModel - Whether to wait for the model to be ready before sending the request. Default: false
- api
Version string - Cohere API version.
v1uses the legacy/v1/chatendpoint;v2(default) uses/v2/chatand supports tool calling. possible known values include one of ["v1", "v2"]; Default: "v2" - cache
Read numberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write numberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write AiCost Lists Gateway Model Model Target Config Cohere Cache Write Cost List[] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window AiFactors Gateway Model Model Target Config Cohere Context Window Factor[] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embedding
Input stringType - The intended downstream use of the embeddings to improve model quality. possible known values include one of ["classification", "clustering", "image", "searchdocument", "searchquery"]; Default: "classification"
- embeddings
Dimensions number - The number of dimensions for embedding outputs.
- input
Cost number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens number - The maximum number of tokens to generate in the response.
- output
Cost number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier AiFactors Gateway Model Model Target Config Cohere Service Tier Factor[] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K number - Limits the number of highest-probability tokens considered during generation.
- top
P number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url string - The upstream URL for the model endpoint.
- wait
For booleanModel - Whether to wait for the model to be ready before sending the request. Default: false
- api_
version str - Cohere API version.
v1uses the legacy/v1/chatendpoint;v2(default) uses/v2/chatand supports tool calling. possible known values include one of ["v1", "v2"]; Default: "v2" - cache_
read_ floatcost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ floatcost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ Sequence[Aicost_ lists Gateway Model Model Target Config Cohere Cache Write Cost List] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ Sequence[Aifactors Gateway Model Model Target Config Cohere Context Window Factor] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embedding_
input_ strtype - The intended downstream use of the embeddings to improve model quality. possible known values include one of ["classification", "clustering", "image", "searchdocument", "searchquery"]; Default: "classification"
- embeddings_
dimensions float - The number of dimensions for embedding outputs.
- input_
cost float - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens float - The maximum number of tokens to generate in the response.
- output_
cost float - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ Sequence[Aifactors Gateway Model Model Target Config Cohere Service Tier Factor] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature float
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k float - Limits the number of highest-probability tokens considered during generation.
- top_
p float - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url str - The upstream URL for the model endpoint.
- wait_
for_ boolmodel - Whether to wait for the model to be ready before sending the request. Default: false
- api
Version String - Cohere API version.
v1uses the legacy/v1/chatendpoint;v2(default) uses/v2/chatand supports tool calling. possible known values include one of ["v1", "v2"]; Default: "v2" - cache
Read NumberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write NumberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<Property Map>Cost Lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<Property Map>Factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embedding
Input StringType - The intended downstream use of the embeddings to improve model quality. possible known values include one of ["classification", "clustering", "image", "searchdocument", "searchquery"]; Default: "classification"
- embeddings
Dimensions Number - The number of dimensions for embedding outputs.
- input
Cost Number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Number - The maximum number of tokens to generate in the response.
- output
Cost Number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<Property Map>Factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Number - Limits the number of highest-probability tokens considered during generation.
- top
P Number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
- wait
For BooleanModel - Whether to wait for the model to be ready before sending the request. Default: false
AiGatewayModelModelTargetConfigCohereCacheWriteCostList, AiGatewayModelModelTargetConfigCohereCacheWriteCostListArgs
AiGatewayModelModelTargetConfigCohereContextWindowFactor, AiGatewayModelModelTargetConfigCohereContextWindowFactorArgs
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor double - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor double - Multiplier applied to output pricing above the threshold. Not Null
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor float64 - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor float64 - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor number - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor number - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Double - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Double - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor number - Multiplier applied to output pricing above the threshold. Not Null
- above str
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor float - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor float - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Number - Multiplier applied to output pricing above the threshold. Not Null
AiGatewayModelModelTargetConfigCohereServiceTierFactor, AiGatewayModelModelTargetConfigCohereServiceTierFactorArgs
- Factor double
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- Factor float64
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Double
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor float
- Multiplier applied to the whole request for this service tier. Not Null
- tier str
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Number
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
AiGatewayModelModelTargetConfigDashscope, AiGatewayModelModelTargetConfigDashscopeArgs
- Cache
Read doubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write doubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write List<AiCost Lists Gateway Model Model Target Config Dashscope Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window List<AiFactors Gateway Model Model Target Config Dashscope Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions double - The number of dimensions for embedding outputs.
- Input
Cost double - Cost per 1M input tokens for billing and cost tracking.
- International bool
- Whether to use the international DashScope endpoint. Default: true
- Max
Tokens double - The maximum number of tokens to generate in the response.
- Output
Cost double - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier List<AiFactors Gateway Model Model Target Config Dashscope Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature double
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K double - Limits the number of highest-probability tokens considered during generation.
- Top
P double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- Cache
Read float64Cost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write float64Cost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write []AiCost Lists Gateway Model Model Target Config Dashscope Cache Write Cost List - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window []AiFactors Gateway Model Model Target Config Dashscope Context Window Factor - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions float64 - The number of dimensions for embedding outputs.
- Input
Cost float64 - Cost per 1M input tokens for billing and cost tracking.
- International bool
- Whether to use the international DashScope endpoint. Default: true
- Max
Tokens float64 - The maximum number of tokens to generate in the response.
- Output
Cost float64 - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier []AiFactors Gateway Model Model Target Config Dashscope Service Tier Factor - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature float64
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K float64 - Limits the number of highest-probability tokens considered during generation.
- Top
P float64 - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- cache_
read_ numbercost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ numbercost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ list(object)cost_ lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ list(object)factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions number - The number of dimensions for embedding outputs.
- input_
cost number - Cost per 1M input tokens for billing and cost tracking.
- international bool
- Whether to use the international DashScope endpoint. Default: true
- max_
tokens number - The maximum number of tokens to generate in the response.
- output_
cost number - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ list(object)factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k number - Limits the number of highest-probability tokens considered during generation.
- top_
p number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url string - The upstream URL for the model endpoint.
- cache
Read DoubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write DoubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<AiCost Lists Gateway Model Model Target Config Dashscope Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<AiFactors Gateway Model Model Target Config Dashscope Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Double - The number of dimensions for embedding outputs.
- input
Cost Double - Cost per 1M input tokens for billing and cost tracking.
- international Boolean
- Whether to use the international DashScope endpoint. Default: true
- max
Tokens Double - The maximum number of tokens to generate in the response.
- output
Cost Double - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<AiFactors Gateway Model Model Target Config Dashscope Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Double
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Double - Limits the number of highest-probability tokens considered during generation.
- top
P Double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
- cache
Read numberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write numberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write AiCost Lists Gateway Model Model Target Config Dashscope Cache Write Cost List[] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window AiFactors Gateway Model Model Target Config Dashscope Context Window Factor[] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions number - The number of dimensions for embedding outputs.
- input
Cost number - Cost per 1M input tokens for billing and cost tracking.
- international boolean
- Whether to use the international DashScope endpoint. Default: true
- max
Tokens number - The maximum number of tokens to generate in the response.
- output
Cost number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier AiFactors Gateway Model Model Target Config Dashscope Service Tier Factor[] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K number - Limits the number of highest-probability tokens considered during generation.
- top
P number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url string - The upstream URL for the model endpoint.
- cache_
read_ floatcost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ floatcost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ Sequence[Aicost_ lists Gateway Model Model Target Config Dashscope Cache Write Cost List] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ Sequence[Aifactors Gateway Model Model Target Config Dashscope Context Window Factor] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions float - The number of dimensions for embedding outputs.
- input_
cost float - Cost per 1M input tokens for billing and cost tracking.
- international bool
- Whether to use the international DashScope endpoint. Default: true
- max_
tokens float - The maximum number of tokens to generate in the response.
- output_
cost float - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ Sequence[Aifactors Gateway Model Model Target Config Dashscope Service Tier Factor] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature float
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k float - Limits the number of highest-probability tokens considered during generation.
- top_
p float - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url str - The upstream URL for the model endpoint.
- cache
Read NumberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write NumberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<Property Map>Cost Lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<Property Map>Factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Number - The number of dimensions for embedding outputs.
- input
Cost Number - Cost per 1M input tokens for billing and cost tracking.
- international Boolean
- Whether to use the international DashScope endpoint. Default: true
- max
Tokens Number - The maximum number of tokens to generate in the response.
- output
Cost Number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<Property Map>Factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Number - Limits the number of highest-probability tokens considered during generation.
- top
P Number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
AiGatewayModelModelTargetConfigDashscopeCacheWriteCostList, AiGatewayModelModelTargetConfigDashscopeCacheWriteCostListArgs
AiGatewayModelModelTargetConfigDashscopeContextWindowFactor, AiGatewayModelModelTargetConfigDashscopeContextWindowFactorArgs
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor double - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor double - Multiplier applied to output pricing above the threshold. Not Null
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor float64 - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor float64 - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor number - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor number - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Double - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Double - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor number - Multiplier applied to output pricing above the threshold. Not Null
- above str
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor float - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor float - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Number - Multiplier applied to output pricing above the threshold. Not Null
AiGatewayModelModelTargetConfigDashscopeServiceTierFactor, AiGatewayModelModelTargetConfigDashscopeServiceTierFactorArgs
- Factor double
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- Factor float64
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Double
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor float
- Multiplier applied to the whole request for this service tier. Not Null
- tier str
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Number
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
AiGatewayModelModelTargetConfigDatabricks, AiGatewayModelModelTargetConfigDatabricksArgs
- Cache
Read doubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write doubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write List<AiCost Lists Gateway Model Model Target Config Databricks Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window List<AiFactors Gateway Model Model Target Config Databricks Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions double - The number of dimensions for embedding outputs.
- Input
Cost double - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens double - The maximum number of tokens to generate in the response.
- Output
Cost double - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier List<AiFactors Gateway Model Model Target Config Databricks Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature double
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K double - Limits the number of highest-probability tokens considered during generation.
- Top
P double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- Workspace
Instance stringId - The Databricks workspace instance ID. Not Null
- Cache
Read float64Cost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write float64Cost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write []AiCost Lists Gateway Model Model Target Config Databricks Cache Write Cost List - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window []AiFactors Gateway Model Model Target Config Databricks Context Window Factor - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions float64 - The number of dimensions for embedding outputs.
- Input
Cost float64 - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens float64 - The maximum number of tokens to generate in the response.
- Output
Cost float64 - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier []AiFactors Gateway Model Model Target Config Databricks Service Tier Factor - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature float64
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K float64 - Limits the number of highest-probability tokens considered during generation.
- Top
P float64 - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- Workspace
Instance stringId - The Databricks workspace instance ID. Not Null
- cache_
read_ numbercost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ numbercost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ list(object)cost_ lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ list(object)factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions number - The number of dimensions for embedding outputs.
- input_
cost number - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens number - The maximum number of tokens to generate in the response.
- output_
cost number - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ list(object)factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k number - Limits the number of highest-probability tokens considered during generation.
- top_
p number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url string - The upstream URL for the model endpoint.
- workspace_
instance_ stringid - The Databricks workspace instance ID. Not Null
- cache
Read DoubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write DoubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<AiCost Lists Gateway Model Model Target Config Databricks Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<AiFactors Gateway Model Model Target Config Databricks Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Double - The number of dimensions for embedding outputs.
- input
Cost Double - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Double - The maximum number of tokens to generate in the response.
- output
Cost Double - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<AiFactors Gateway Model Model Target Config Databricks Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Double
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Double - Limits the number of highest-probability tokens considered during generation.
- top
P Double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
- workspace
Instance StringId - The Databricks workspace instance ID. Not Null
- cache
Read numberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write numberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write AiCost Lists Gateway Model Model Target Config Databricks Cache Write Cost List[] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window AiFactors Gateway Model Model Target Config Databricks Context Window Factor[] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions number - The number of dimensions for embedding outputs.
- input
Cost number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens number - The maximum number of tokens to generate in the response.
- output
Cost number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier AiFactors Gateway Model Model Target Config Databricks Service Tier Factor[] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K number - Limits the number of highest-probability tokens considered during generation.
- top
P number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url string - The upstream URL for the model endpoint.
- workspace
Instance stringId - The Databricks workspace instance ID. Not Null
- cache_
read_ floatcost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ floatcost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ Sequence[Aicost_ lists Gateway Model Model Target Config Databricks Cache Write Cost List] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ Sequence[Aifactors Gateway Model Model Target Config Databricks Context Window Factor] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions float - The number of dimensions for embedding outputs.
- input_
cost float - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens float - The maximum number of tokens to generate in the response.
- output_
cost float - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ Sequence[Aifactors Gateway Model Model Target Config Databricks Service Tier Factor] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature float
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k float - Limits the number of highest-probability tokens considered during generation.
- top_
p float - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url str - The upstream URL for the model endpoint.
- workspace_
instance_ strid - The Databricks workspace instance ID. Not Null
- cache
Read NumberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write NumberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<Property Map>Cost Lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<Property Map>Factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Number - The number of dimensions for embedding outputs.
- input
Cost Number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Number - The maximum number of tokens to generate in the response.
- output
Cost Number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<Property Map>Factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Number - Limits the number of highest-probability tokens considered during generation.
- top
P Number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
- workspace
Instance StringId - The Databricks workspace instance ID. Not Null
AiGatewayModelModelTargetConfigDatabricksCacheWriteCostList, AiGatewayModelModelTargetConfigDatabricksCacheWriteCostListArgs
AiGatewayModelModelTargetConfigDatabricksContextWindowFactor, AiGatewayModelModelTargetConfigDatabricksContextWindowFactorArgs
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor double - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor double - Multiplier applied to output pricing above the threshold. Not Null
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor float64 - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor float64 - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor number - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor number - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Double - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Double - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor number - Multiplier applied to output pricing above the threshold. Not Null
- above str
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor float - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor float - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Number - Multiplier applied to output pricing above the threshold. Not Null
AiGatewayModelModelTargetConfigDatabricksServiceTierFactor, AiGatewayModelModelTargetConfigDatabricksServiceTierFactorArgs
- Factor double
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- Factor float64
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Double
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor float
- Multiplier applied to the whole request for this service tier. Not Null
- tier str
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Number
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
AiGatewayModelModelTargetConfigDeepseek, AiGatewayModelModelTargetConfigDeepseekArgs
- Cache
Read doubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write doubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write List<AiCost Lists Gateway Model Model Target Config Deepseek Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window List<AiFactors Gateway Model Model Target Config Deepseek Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions double - The number of dimensions for embedding outputs.
- Input
Cost double - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens double - The maximum number of tokens to generate in the response.
- Output
Cost double - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier List<AiFactors Gateway Model Model Target Config Deepseek Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature double
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K double - Limits the number of highest-probability tokens considered during generation.
- Top
P double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- Cache
Read float64Cost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write float64Cost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write []AiCost Lists Gateway Model Model Target Config Deepseek Cache Write Cost List - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window []AiFactors Gateway Model Model Target Config Deepseek Context Window Factor - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions float64 - The number of dimensions for embedding outputs.
- Input
Cost float64 - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens float64 - The maximum number of tokens to generate in the response.
- Output
Cost float64 - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier []AiFactors Gateway Model Model Target Config Deepseek Service Tier Factor - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature float64
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K float64 - Limits the number of highest-probability tokens considered during generation.
- Top
P float64 - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- cache_
read_ numbercost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ numbercost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ list(object)cost_ lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ list(object)factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions number - The number of dimensions for embedding outputs.
- input_
cost number - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens number - The maximum number of tokens to generate in the response.
- output_
cost number - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ list(object)factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k number - Limits the number of highest-probability tokens considered during generation.
- top_
p number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url string - The upstream URL for the model endpoint.
- cache
Read DoubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write DoubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<AiCost Lists Gateway Model Model Target Config Deepseek Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<AiFactors Gateway Model Model Target Config Deepseek Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Double - The number of dimensions for embedding outputs.
- input
Cost Double - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Double - The maximum number of tokens to generate in the response.
- output
Cost Double - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<AiFactors Gateway Model Model Target Config Deepseek Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Double
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Double - Limits the number of highest-probability tokens considered during generation.
- top
P Double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
- cache
Read numberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write numberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write AiCost Lists Gateway Model Model Target Config Deepseek Cache Write Cost List[] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window AiFactors Gateway Model Model Target Config Deepseek Context Window Factor[] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions number - The number of dimensions for embedding outputs.
- input
Cost number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens number - The maximum number of tokens to generate in the response.
- output
Cost number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier AiFactors Gateway Model Model Target Config Deepseek Service Tier Factor[] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K number - Limits the number of highest-probability tokens considered during generation.
- top
P number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url string - The upstream URL for the model endpoint.
- cache_
read_ floatcost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ floatcost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ Sequence[Aicost_ lists Gateway Model Model Target Config Deepseek Cache Write Cost List] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ Sequence[Aifactors Gateway Model Model Target Config Deepseek Context Window Factor] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions float - The number of dimensions for embedding outputs.
- input_
cost float - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens float - The maximum number of tokens to generate in the response.
- output_
cost float - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ Sequence[Aifactors Gateway Model Model Target Config Deepseek Service Tier Factor] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature float
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k float - Limits the number of highest-probability tokens considered during generation.
- top_
p float - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url str - The upstream URL for the model endpoint.
- cache
Read NumberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write NumberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<Property Map>Cost Lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<Property Map>Factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Number - The number of dimensions for embedding outputs.
- input
Cost Number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Number - The maximum number of tokens to generate in the response.
- output
Cost Number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<Property Map>Factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Number - Limits the number of highest-probability tokens considered during generation.
- top
P Number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
AiGatewayModelModelTargetConfigDeepseekCacheWriteCostList, AiGatewayModelModelTargetConfigDeepseekCacheWriteCostListArgs
AiGatewayModelModelTargetConfigDeepseekContextWindowFactor, AiGatewayModelModelTargetConfigDeepseekContextWindowFactorArgs
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor double - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor double - Multiplier applied to output pricing above the threshold. Not Null
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor float64 - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor float64 - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor number - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor number - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Double - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Double - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor number - Multiplier applied to output pricing above the threshold. Not Null
- above str
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor float - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor float - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Number - Multiplier applied to output pricing above the threshold. Not Null
AiGatewayModelModelTargetConfigDeepseekServiceTierFactor, AiGatewayModelModelTargetConfigDeepseekServiceTierFactorArgs
- Factor double
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- Factor float64
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Double
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor float
- Multiplier applied to the whole request for this service tier. Not Null
- tier str
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Number
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
AiGatewayModelModelTargetConfigGemini, AiGatewayModelModelTargetConfigGeminiArgs
- Cache
Read doubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write doubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write List<AiCost Lists Gateway Model Model Target Config Gemini Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window List<AiFactors Gateway Model Model Target Config Gemini Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions double - The number of dimensions for embedding outputs.
- Gcp
Environment AiGateway Model Model Target Config Gemini Gcp Environment - Configuration for a model hosted on Google Cloud Project.
- Input
Cost double - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens double - The maximum number of tokens to generate in the response.
- Output
Cost double - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier List<AiFactors Gateway Model Model Target Config Gemini Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature double
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K double - Limits the number of highest-probability tokens considered during generation.
- Top
P double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- Cache
Read float64Cost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write float64Cost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write []AiCost Lists Gateway Model Model Target Config Gemini Cache Write Cost List - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window []AiFactors Gateway Model Model Target Config Gemini Context Window Factor - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions float64 - The number of dimensions for embedding outputs.
- Gcp
Environment AiGateway Model Model Target Config Gemini Gcp Environment - Configuration for a model hosted on Google Cloud Project.
- Input
Cost float64 - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens float64 - The maximum number of tokens to generate in the response.
- Output
Cost float64 - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier []AiFactors Gateway Model Model Target Config Gemini Service Tier Factor - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature float64
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K float64 - Limits the number of highest-probability tokens considered during generation.
- Top
P float64 - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- cache_
read_ numbercost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ numbercost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ list(object)cost_ lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ list(object)factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions number - The number of dimensions for embedding outputs.
- gcp_
environment object - Configuration for a model hosted on Google Cloud Project.
- input_
cost number - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens number - The maximum number of tokens to generate in the response.
- output_
cost number - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ list(object)factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k number - Limits the number of highest-probability tokens considered during generation.
- top_
p number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url string - The upstream URL for the model endpoint.
- cache
Read DoubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write DoubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<AiCost Lists Gateway Model Model Target Config Gemini Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<AiFactors Gateway Model Model Target Config Gemini Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Double - The number of dimensions for embedding outputs.
- gcp
Environment AiGateway Model Model Target Config Gemini Gcp Environment - Configuration for a model hosted on Google Cloud Project.
- input
Cost Double - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Double - The maximum number of tokens to generate in the response.
- output
Cost Double - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<AiFactors Gateway Model Model Target Config Gemini Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Double
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Double - Limits the number of highest-probability tokens considered during generation.
- top
P Double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
- cache
Read numberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write numberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write AiCost Lists Gateway Model Model Target Config Gemini Cache Write Cost List[] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window AiFactors Gateway Model Model Target Config Gemini Context Window Factor[] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions number - The number of dimensions for embedding outputs.
- gcp
Environment AiGateway Model Model Target Config Gemini Gcp Environment - Configuration for a model hosted on Google Cloud Project.
- input
Cost number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens number - The maximum number of tokens to generate in the response.
- output
Cost number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier AiFactors Gateway Model Model Target Config Gemini Service Tier Factor[] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K number - Limits the number of highest-probability tokens considered during generation.
- top
P number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url string - The upstream URL for the model endpoint.
- cache_
read_ floatcost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ floatcost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ Sequence[Aicost_ lists Gateway Model Model Target Config Gemini Cache Write Cost List] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ Sequence[Aifactors Gateway Model Model Target Config Gemini Context Window Factor] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions float - The number of dimensions for embedding outputs.
- gcp_
environment AiGateway Model Model Target Config Gemini Gcp Environment - Configuration for a model hosted on Google Cloud Project.
- input_
cost float - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens float - The maximum number of tokens to generate in the response.
- output_
cost float - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ Sequence[Aifactors Gateway Model Model Target Config Gemini Service Tier Factor] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature float
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k float - Limits the number of highest-probability tokens considered during generation.
- top_
p float - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url str - The upstream URL for the model endpoint.
- cache
Read NumberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write NumberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<Property Map>Cost Lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<Property Map>Factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Number - The number of dimensions for embedding outputs.
- gcp
Environment Property Map - Configuration for a model hosted on Google Cloud Project.
- input
Cost Number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Number - The maximum number of tokens to generate in the response.
- output
Cost Number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<Property Map>Factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Number - Limits the number of highest-probability tokens considered during generation.
- top
P Number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
AiGatewayModelModelTargetConfigGeminiCacheWriteCostList, AiGatewayModelModelTargetConfigGeminiCacheWriteCostListArgs
AiGatewayModelModelTargetConfigGeminiContextWindowFactor, AiGatewayModelModelTargetConfigGeminiContextWindowFactorArgs
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor double - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor double - Multiplier applied to output pricing above the threshold. Not Null
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor float64 - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor float64 - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor number - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor number - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Double - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Double - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor number - Multiplier applied to output pricing above the threshold. Not Null
- above str
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor float - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor float - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Number - Multiplier applied to output pricing above the threshold. Not Null
AiGatewayModelModelTargetConfigGeminiGcpEnvironment, AiGatewayModelModelTargetConfigGeminiGcpEnvironmentArgs
- Api
Endpoint string - The custom API endpoint for the Gemini model. Not Null
- Location
Id string - The Google Cloud location ID for the model endpoint. Not Null
- Project
Id string - The Google Cloud project ID for the model endpoint. Not Null
- Api
Endpoint string - The custom API endpoint for the Gemini model. Not Null
- Location
Id string - The Google Cloud location ID for the model endpoint. Not Null
- Project
Id string - The Google Cloud project ID for the model endpoint. Not Null
- api_
endpoint string - The custom API endpoint for the Gemini model. Not Null
- location_
id string - The Google Cloud location ID for the model endpoint. Not Null
- project_
id string - The Google Cloud project ID for the model endpoint. Not Null
- api
Endpoint String - The custom API endpoint for the Gemini model. Not Null
- location
Id String - The Google Cloud location ID for the model endpoint. Not Null
- project
Id String - The Google Cloud project ID for the model endpoint. Not Null
- api
Endpoint string - The custom API endpoint for the Gemini model. Not Null
- location
Id string - The Google Cloud location ID for the model endpoint. Not Null
- project
Id string - The Google Cloud project ID for the model endpoint. Not Null
- api_
endpoint str - The custom API endpoint for the Gemini model. Not Null
- location_
id str - The Google Cloud location ID for the model endpoint. Not Null
- project_
id str - The Google Cloud project ID for the model endpoint. Not Null
- api
Endpoint String - The custom API endpoint for the Gemini model. Not Null
- location
Id String - The Google Cloud location ID for the model endpoint. Not Null
- project
Id String - The Google Cloud project ID for the model endpoint. Not Null
AiGatewayModelModelTargetConfigGeminiServiceTierFactor, AiGatewayModelModelTargetConfigGeminiServiceTierFactorArgs
- Factor double
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- Factor float64
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Double
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor float
- Multiplier applied to the whole request for this service tier. Not Null
- tier str
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Number
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
AiGatewayModelModelTargetConfigHuggingface, AiGatewayModelModelTargetConfigHuggingfaceArgs
- Cache
Read doubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write doubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write List<AiCost Lists Gateway Model Model Target Config Huggingface Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window List<AiFactors Gateway Model Model Target Config Huggingface Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions double - The number of dimensions for embedding outputs.
- Input
Cost double - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens double - The maximum number of tokens to generate in the response.
- Output
Cost double - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier List<AiFactors Gateway Model Model Target Config Huggingface Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature double
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K double - Limits the number of highest-probability tokens considered during generation.
- Top
P double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- Use
Cache bool - Whether to use the Hugging Face inference cache. Default: false
- Wait
For boolModel - Whether to wait for the model to load if it is not ready. Default: false
- Cache
Read float64Cost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write float64Cost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write []AiCost Lists Gateway Model Model Target Config Huggingface Cache Write Cost List - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window []AiFactors Gateway Model Model Target Config Huggingface Context Window Factor - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions float64 - The number of dimensions for embedding outputs.
- Input
Cost float64 - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens float64 - The maximum number of tokens to generate in the response.
- Output
Cost float64 - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier []AiFactors Gateway Model Model Target Config Huggingface Service Tier Factor - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature float64
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K float64 - Limits the number of highest-probability tokens considered during generation.
- Top
P float64 - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- Use
Cache bool - Whether to use the Hugging Face inference cache. Default: false
- Wait
For boolModel - Whether to wait for the model to load if it is not ready. Default: false
- cache_
read_ numbercost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ numbercost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ list(object)cost_ lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ list(object)factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions number - The number of dimensions for embedding outputs.
- input_
cost number - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens number - The maximum number of tokens to generate in the response.
- output_
cost number - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ list(object)factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k number - Limits the number of highest-probability tokens considered during generation.
- top_
p number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url string - The upstream URL for the model endpoint.
- use_
cache bool - Whether to use the Hugging Face inference cache. Default: false
- wait_
for_ boolmodel - Whether to wait for the model to load if it is not ready. Default: false
- cache
Read DoubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write DoubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<AiCost Lists Gateway Model Model Target Config Huggingface Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<AiFactors Gateway Model Model Target Config Huggingface Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Double - The number of dimensions for embedding outputs.
- input
Cost Double - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Double - The maximum number of tokens to generate in the response.
- output
Cost Double - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<AiFactors Gateway Model Model Target Config Huggingface Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Double
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Double - Limits the number of highest-probability tokens considered during generation.
- top
P Double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
- use
Cache Boolean - Whether to use the Hugging Face inference cache. Default: false
- wait
For BooleanModel - Whether to wait for the model to load if it is not ready. Default: false
- cache
Read numberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write numberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write AiCost Lists Gateway Model Model Target Config Huggingface Cache Write Cost List[] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window AiFactors Gateway Model Model Target Config Huggingface Context Window Factor[] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions number - The number of dimensions for embedding outputs.
- input
Cost number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens number - The maximum number of tokens to generate in the response.
- output
Cost number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier AiFactors Gateway Model Model Target Config Huggingface Service Tier Factor[] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K number - Limits the number of highest-probability tokens considered during generation.
- top
P number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url string - The upstream URL for the model endpoint.
- use
Cache boolean - Whether to use the Hugging Face inference cache. Default: false
- wait
For booleanModel - Whether to wait for the model to load if it is not ready. Default: false
- cache_
read_ floatcost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ floatcost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ Sequence[Aicost_ lists Gateway Model Model Target Config Huggingface Cache Write Cost List] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ Sequence[Aifactors Gateway Model Model Target Config Huggingface Context Window Factor] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions float - The number of dimensions for embedding outputs.
- input_
cost float - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens float - The maximum number of tokens to generate in the response.
- output_
cost float - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ Sequence[Aifactors Gateway Model Model Target Config Huggingface Service Tier Factor] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature float
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k float - Limits the number of highest-probability tokens considered during generation.
- top_
p float - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url str - The upstream URL for the model endpoint.
- use_
cache bool - Whether to use the Hugging Face inference cache. Default: false
- wait_
for_ boolmodel - Whether to wait for the model to load if it is not ready. Default: false
- cache
Read NumberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write NumberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<Property Map>Cost Lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<Property Map>Factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Number - The number of dimensions for embedding outputs.
- input
Cost Number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Number - The maximum number of tokens to generate in the response.
- output
Cost Number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<Property Map>Factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Number - Limits the number of highest-probability tokens considered during generation.
- top
P Number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
- use
Cache Boolean - Whether to use the Hugging Face inference cache. Default: false
- wait
For BooleanModel - Whether to wait for the model to load if it is not ready. Default: false
AiGatewayModelModelTargetConfigHuggingfaceCacheWriteCostList, AiGatewayModelModelTargetConfigHuggingfaceCacheWriteCostListArgs
AiGatewayModelModelTargetConfigHuggingfaceContextWindowFactor, AiGatewayModelModelTargetConfigHuggingfaceContextWindowFactorArgs
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor double - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor double - Multiplier applied to output pricing above the threshold. Not Null
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor float64 - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor float64 - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor number - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor number - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Double - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Double - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor number - Multiplier applied to output pricing above the threshold. Not Null
- above str
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor float - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor float - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Number - Multiplier applied to output pricing above the threshold. Not Null
AiGatewayModelModelTargetConfigHuggingfaceServiceTierFactor, AiGatewayModelModelTargetConfigHuggingfaceServiceTierFactorArgs
- Factor double
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- Factor float64
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Double
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor float
- Multiplier applied to the whole request for this service tier. Not Null
- tier str
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Number
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
AiGatewayModelModelTargetConfigKimi, AiGatewayModelModelTargetConfigKimiArgs
- Cache
Read doubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write doubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write List<AiCost Lists Gateway Model Model Target Config Kimi Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window List<AiFactors Gateway Model Model Target Config Kimi Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions double - The number of dimensions for embedding outputs.
- Input
Cost double - Cost per 1M input tokens for billing and cost tracking.
- International bool
- When
true, requests are sent toapi.moonshot.ai(international). Whenfalse, requests are sent toapi.moonshot.cn(mainland China). Default: true - Max
Tokens double - The maximum number of tokens to generate in the response.
- Output
Cost double - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier List<AiFactors Gateway Model Model Target Config Kimi Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature double
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K double - Limits the number of highest-probability tokens considered during generation.
- Top
P double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- Cache
Read float64Cost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write float64Cost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write []AiCost Lists Gateway Model Model Target Config Kimi Cache Write Cost List - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window []AiFactors Gateway Model Model Target Config Kimi Context Window Factor - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions float64 - The number of dimensions for embedding outputs.
- Input
Cost float64 - Cost per 1M input tokens for billing and cost tracking.
- International bool
- When
true, requests are sent toapi.moonshot.ai(international). Whenfalse, requests are sent toapi.moonshot.cn(mainland China). Default: true - Max
Tokens float64 - The maximum number of tokens to generate in the response.
- Output
Cost float64 - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier []AiFactors Gateway Model Model Target Config Kimi Service Tier Factor - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature float64
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K float64 - Limits the number of highest-probability tokens considered during generation.
- Top
P float64 - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- cache_
read_ numbercost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ numbercost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ list(object)cost_ lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ list(object)factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions number - The number of dimensions for embedding outputs.
- input_
cost number - Cost per 1M input tokens for billing and cost tracking.
- international bool
- When
true, requests are sent toapi.moonshot.ai(international). Whenfalse, requests are sent toapi.moonshot.cn(mainland China). Default: true - max_
tokens number - The maximum number of tokens to generate in the response.
- output_
cost number - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ list(object)factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k number - Limits the number of highest-probability tokens considered during generation.
- top_
p number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url string - The upstream URL for the model endpoint.
- cache
Read DoubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write DoubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<AiCost Lists Gateway Model Model Target Config Kimi Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<AiFactors Gateway Model Model Target Config Kimi Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Double - The number of dimensions for embedding outputs.
- input
Cost Double - Cost per 1M input tokens for billing and cost tracking.
- international Boolean
- When
true, requests are sent toapi.moonshot.ai(international). Whenfalse, requests are sent toapi.moonshot.cn(mainland China). Default: true - max
Tokens Double - The maximum number of tokens to generate in the response.
- output
Cost Double - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<AiFactors Gateway Model Model Target Config Kimi Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Double
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Double - Limits the number of highest-probability tokens considered during generation.
- top
P Double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
- cache
Read numberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write numberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write AiCost Lists Gateway Model Model Target Config Kimi Cache Write Cost List[] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window AiFactors Gateway Model Model Target Config Kimi Context Window Factor[] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions number - The number of dimensions for embedding outputs.
- input
Cost number - Cost per 1M input tokens for billing and cost tracking.
- international boolean
- When
true, requests are sent toapi.moonshot.ai(international). Whenfalse, requests are sent toapi.moonshot.cn(mainland China). Default: true - max
Tokens number - The maximum number of tokens to generate in the response.
- output
Cost number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier AiFactors Gateway Model Model Target Config Kimi Service Tier Factor[] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K number - Limits the number of highest-probability tokens considered during generation.
- top
P number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url string - The upstream URL for the model endpoint.
- cache_
read_ floatcost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ floatcost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ Sequence[Aicost_ lists Gateway Model Model Target Config Kimi Cache Write Cost List] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ Sequence[Aifactors Gateway Model Model Target Config Kimi Context Window Factor] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions float - The number of dimensions for embedding outputs.
- input_
cost float - Cost per 1M input tokens for billing and cost tracking.
- international bool
- When
true, requests are sent toapi.moonshot.ai(international). Whenfalse, requests are sent toapi.moonshot.cn(mainland China). Default: true - max_
tokens float - The maximum number of tokens to generate in the response.
- output_
cost float - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ Sequence[Aifactors Gateway Model Model Target Config Kimi Service Tier Factor] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature float
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k float - Limits the number of highest-probability tokens considered during generation.
- top_
p float - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url str - The upstream URL for the model endpoint.
- cache
Read NumberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write NumberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<Property Map>Cost Lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<Property Map>Factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Number - The number of dimensions for embedding outputs.
- input
Cost Number - Cost per 1M input tokens for billing and cost tracking.
- international Boolean
- When
true, requests are sent toapi.moonshot.ai(international). Whenfalse, requests are sent toapi.moonshot.cn(mainland China). Default: true - max
Tokens Number - The maximum number of tokens to generate in the response.
- output
Cost Number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<Property Map>Factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Number - Limits the number of highest-probability tokens considered during generation.
- top
P Number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
AiGatewayModelModelTargetConfigKimiCacheWriteCostList, AiGatewayModelModelTargetConfigKimiCacheWriteCostListArgs
AiGatewayModelModelTargetConfigKimiContextWindowFactor, AiGatewayModelModelTargetConfigKimiContextWindowFactorArgs
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor double - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor double - Multiplier applied to output pricing above the threshold. Not Null
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor float64 - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor float64 - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor number - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor number - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Double - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Double - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor number - Multiplier applied to output pricing above the threshold. Not Null
- above str
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor float - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor float - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Number - Multiplier applied to output pricing above the threshold. Not Null
AiGatewayModelModelTargetConfigKimiServiceTierFactor, AiGatewayModelModelTargetConfigKimiServiceTierFactorArgs
- Factor double
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- Factor float64
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Double
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor float
- Multiplier applied to the whole request for this service tier. Not Null
- tier str
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Number
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
AiGatewayModelModelTargetConfigLlama2, AiGatewayModelModelTargetConfigLlama2Args
- Cache
Read doubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write doubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write List<AiCost Lists Gateway Model Model Target Config Llama2Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window List<AiFactors Gateway Model Model Target Config Llama2Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions double - The number of dimensions for embedding outputs.
- Format string
- The request format to use when communicating with the Llama2 model. possible known values include one of ["ollama", "openai", "raw"]; Not Null
- Input
Cost double - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens double - The maximum number of tokens to generate in the response.
- Output
Cost double - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier List<AiFactors Gateway Model Model Target Config Llama2Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature double
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K double - Limits the number of highest-probability tokens considered during generation.
- Top
P double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint. Not Null
- Cache
Read float64Cost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write float64Cost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write []AiCost Lists Gateway Model Model Target Config Llama2Cache Write Cost List - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window []AiFactors Gateway Model Model Target Config Llama2Context Window Factor - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions float64 - The number of dimensions for embedding outputs.
- Format string
- The request format to use when communicating with the Llama2 model. possible known values include one of ["ollama", "openai", "raw"]; Not Null
- Input
Cost float64 - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens float64 - The maximum number of tokens to generate in the response.
- Output
Cost float64 - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier []AiFactors Gateway Model Model Target Config Llama2Service Tier Factor - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature float64
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K float64 - Limits the number of highest-probability tokens considered during generation.
- Top
P float64 - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint. Not Null
- cache_
read_ numbercost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ numbercost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ list(object)cost_ lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ list(object)factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions number - The number of dimensions for embedding outputs.
- format string
- The request format to use when communicating with the Llama2 model. possible known values include one of ["ollama", "openai", "raw"]; Not Null
- input_
cost number - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens number - The maximum number of tokens to generate in the response.
- output_
cost number - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ list(object)factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k number - Limits the number of highest-probability tokens considered during generation.
- top_
p number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url string - The upstream URL for the model endpoint. Not Null
- cache
Read DoubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write DoubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<AiCost Lists Gateway Model Model Target Config Llama2Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<AiFactors Gateway Model Model Target Config Llama2Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Double - The number of dimensions for embedding outputs.
- format String
- The request format to use when communicating with the Llama2 model. possible known values include one of ["ollama", "openai", "raw"]; Not Null
- input
Cost Double - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Double - The maximum number of tokens to generate in the response.
- output
Cost Double - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<AiFactors Gateway Model Model Target Config Llama2Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Double
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Double - Limits the number of highest-probability tokens considered during generation.
- top
P Double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint. Not Null
- cache
Read numberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write numberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write AiCost Lists Gateway Model Model Target Config Llama2Cache Write Cost List[] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window AiFactors Gateway Model Model Target Config Llama2Context Window Factor[] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions number - The number of dimensions for embedding outputs.
- format string
- The request format to use when communicating with the Llama2 model. possible known values include one of ["ollama", "openai", "raw"]; Not Null
- input
Cost number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens number - The maximum number of tokens to generate in the response.
- output
Cost number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier AiFactors Gateway Model Model Target Config Llama2Service Tier Factor[] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K number - Limits the number of highest-probability tokens considered during generation.
- top
P number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url string - The upstream URL for the model endpoint. Not Null
- cache_
read_ floatcost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ floatcost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ Sequence[Aicost_ lists Gateway Model Model Target Config Llama2Cache Write Cost List] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ Sequence[Aifactors Gateway Model Model Target Config Llama2Context Window Factor] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions float - The number of dimensions for embedding outputs.
- format str
- The request format to use when communicating with the Llama2 model. possible known values include one of ["ollama", "openai", "raw"]; Not Null
- input_
cost float - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens float - The maximum number of tokens to generate in the response.
- output_
cost float - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ Sequence[Aifactors Gateway Model Model Target Config Llama2Service Tier Factor] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature float
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k float - Limits the number of highest-probability tokens considered during generation.
- top_
p float - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url str - The upstream URL for the model endpoint. Not Null
- cache
Read NumberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write NumberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<Property Map>Cost Lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<Property Map>Factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Number - The number of dimensions for embedding outputs.
- format String
- The request format to use when communicating with the Llama2 model. possible known values include one of ["ollama", "openai", "raw"]; Not Null
- input
Cost Number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Number - The maximum number of tokens to generate in the response.
- output
Cost Number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<Property Map>Factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Number - Limits the number of highest-probability tokens considered during generation.
- top
P Number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint. Not Null
AiGatewayModelModelTargetConfigLlama2CacheWriteCostList, AiGatewayModelModelTargetConfigLlama2CacheWriteCostListArgs
AiGatewayModelModelTargetConfigLlama2ContextWindowFactor, AiGatewayModelModelTargetConfigLlama2ContextWindowFactorArgs
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor double - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor double - Multiplier applied to output pricing above the threshold. Not Null
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor float64 - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor float64 - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor number - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor number - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Double - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Double - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor number - Multiplier applied to output pricing above the threshold. Not Null
- above str
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor float - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor float - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Number - Multiplier applied to output pricing above the threshold. Not Null
AiGatewayModelModelTargetConfigLlama2ServiceTierFactor, AiGatewayModelModelTargetConfigLlama2ServiceTierFactorArgs
- Factor double
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- Factor float64
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Double
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor float
- Multiplier applied to the whole request for this service tier. Not Null
- tier str
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Number
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
AiGatewayModelModelTargetConfigMistral, AiGatewayModelModelTargetConfigMistralArgs
- Cache
Read doubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write doubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write List<AiCost Lists Gateway Model Model Target Config Mistral Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window List<AiFactors Gateway Model Model Target Config Mistral Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions double - The number of dimensions for embedding outputs.
- Format string
- The request format to use when communicating with the Mistral model. possible known values include one of ["ollama", "openai"]; Not Null
- Input
Cost double - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens double - The maximum number of tokens to generate in the response.
- Output
Cost double - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier List<AiFactors Gateway Model Model Target Config Mistral Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature double
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K double - Limits the number of highest-probability tokens considered during generation.
- Top
P double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- Cache
Read float64Cost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write float64Cost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write []AiCost Lists Gateway Model Model Target Config Mistral Cache Write Cost List - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window []AiFactors Gateway Model Model Target Config Mistral Context Window Factor - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions float64 - The number of dimensions for embedding outputs.
- Format string
- The request format to use when communicating with the Mistral model. possible known values include one of ["ollama", "openai"]; Not Null
- Input
Cost float64 - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens float64 - The maximum number of tokens to generate in the response.
- Output
Cost float64 - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier []AiFactors Gateway Model Model Target Config Mistral Service Tier Factor - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature float64
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K float64 - Limits the number of highest-probability tokens considered during generation.
- Top
P float64 - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- cache_
read_ numbercost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ numbercost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ list(object)cost_ lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ list(object)factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions number - The number of dimensions for embedding outputs.
- format string
- The request format to use when communicating with the Mistral model. possible known values include one of ["ollama", "openai"]; Not Null
- input_
cost number - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens number - The maximum number of tokens to generate in the response.
- output_
cost number - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ list(object)factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k number - Limits the number of highest-probability tokens considered during generation.
- top_
p number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url string - The upstream URL for the model endpoint.
- cache
Read DoubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write DoubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<AiCost Lists Gateway Model Model Target Config Mistral Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<AiFactors Gateway Model Model Target Config Mistral Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Double - The number of dimensions for embedding outputs.
- format String
- The request format to use when communicating with the Mistral model. possible known values include one of ["ollama", "openai"]; Not Null
- input
Cost Double - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Double - The maximum number of tokens to generate in the response.
- output
Cost Double - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<AiFactors Gateway Model Model Target Config Mistral Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Double
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Double - Limits the number of highest-probability tokens considered during generation.
- top
P Double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
- cache
Read numberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write numberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write AiCost Lists Gateway Model Model Target Config Mistral Cache Write Cost List[] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window AiFactors Gateway Model Model Target Config Mistral Context Window Factor[] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions number - The number of dimensions for embedding outputs.
- format string
- The request format to use when communicating with the Mistral model. possible known values include one of ["ollama", "openai"]; Not Null
- input
Cost number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens number - The maximum number of tokens to generate in the response.
- output
Cost number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier AiFactors Gateway Model Model Target Config Mistral Service Tier Factor[] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K number - Limits the number of highest-probability tokens considered during generation.
- top
P number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url string - The upstream URL for the model endpoint.
- cache_
read_ floatcost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ floatcost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ Sequence[Aicost_ lists Gateway Model Model Target Config Mistral Cache Write Cost List] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ Sequence[Aifactors Gateway Model Model Target Config Mistral Context Window Factor] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions float - The number of dimensions for embedding outputs.
- format str
- The request format to use when communicating with the Mistral model. possible known values include one of ["ollama", "openai"]; Not Null
- input_
cost float - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens float - The maximum number of tokens to generate in the response.
- output_
cost float - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ Sequence[Aifactors Gateway Model Model Target Config Mistral Service Tier Factor] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature float
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k float - Limits the number of highest-probability tokens considered during generation.
- top_
p float - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url str - The upstream URL for the model endpoint.
- cache
Read NumberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write NumberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<Property Map>Cost Lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<Property Map>Factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Number - The number of dimensions for embedding outputs.
- format String
- The request format to use when communicating with the Mistral model. possible known values include one of ["ollama", "openai"]; Not Null
- input
Cost Number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Number - The maximum number of tokens to generate in the response.
- output
Cost Number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<Property Map>Factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Number - Limits the number of highest-probability tokens considered during generation.
- top
P Number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
AiGatewayModelModelTargetConfigMistralCacheWriteCostList, AiGatewayModelModelTargetConfigMistralCacheWriteCostListArgs
AiGatewayModelModelTargetConfigMistralContextWindowFactor, AiGatewayModelModelTargetConfigMistralContextWindowFactorArgs
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor double - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor double - Multiplier applied to output pricing above the threshold. Not Null
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor float64 - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor float64 - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor number - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor number - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Double - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Double - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor number - Multiplier applied to output pricing above the threshold. Not Null
- above str
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor float - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor float - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Number - Multiplier applied to output pricing above the threshold. Not Null
AiGatewayModelModelTargetConfigMistralServiceTierFactor, AiGatewayModelModelTargetConfigMistralServiceTierFactorArgs
- Factor double
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- Factor float64
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Double
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor float
- Multiplier applied to the whole request for this service tier. Not Null
- tier str
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Number
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
AiGatewayModelModelTargetConfigOllama, AiGatewayModelModelTargetConfigOllamaArgs
- Cache
Read doubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write doubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write List<AiCost Lists Gateway Model Model Target Config Ollama Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window List<AiFactors Gateway Model Model Target Config Ollama Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions double - The number of dimensions for embedding outputs.
- Input
Cost double - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens double - The maximum number of tokens to generate in the response.
- Output
Cost double - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier List<AiFactors Gateway Model Model Target Config Ollama Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature double
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K double - Limits the number of highest-probability tokens considered during generation.
- Top
P double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- Cache
Read float64Cost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write float64Cost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write []AiCost Lists Gateway Model Model Target Config Ollama Cache Write Cost List - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window []AiFactors Gateway Model Model Target Config Ollama Context Window Factor - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions float64 - The number of dimensions for embedding outputs.
- Input
Cost float64 - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens float64 - The maximum number of tokens to generate in the response.
- Output
Cost float64 - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier []AiFactors Gateway Model Model Target Config Ollama Service Tier Factor - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature float64
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K float64 - Limits the number of highest-probability tokens considered during generation.
- Top
P float64 - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- cache_
read_ numbercost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ numbercost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ list(object)cost_ lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ list(object)factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions number - The number of dimensions for embedding outputs.
- input_
cost number - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens number - The maximum number of tokens to generate in the response.
- output_
cost number - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ list(object)factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k number - Limits the number of highest-probability tokens considered during generation.
- top_
p number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url string - The upstream URL for the model endpoint.
- cache
Read DoubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write DoubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<AiCost Lists Gateway Model Model Target Config Ollama Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<AiFactors Gateway Model Model Target Config Ollama Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Double - The number of dimensions for embedding outputs.
- input
Cost Double - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Double - The maximum number of tokens to generate in the response.
- output
Cost Double - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<AiFactors Gateway Model Model Target Config Ollama Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Double
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Double - Limits the number of highest-probability tokens considered during generation.
- top
P Double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
- cache
Read numberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write numberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write AiCost Lists Gateway Model Model Target Config Ollama Cache Write Cost List[] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window AiFactors Gateway Model Model Target Config Ollama Context Window Factor[] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions number - The number of dimensions for embedding outputs.
- input
Cost number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens number - The maximum number of tokens to generate in the response.
- output
Cost number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier AiFactors Gateway Model Model Target Config Ollama Service Tier Factor[] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K number - Limits the number of highest-probability tokens considered during generation.
- top
P number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url string - The upstream URL for the model endpoint.
- cache_
read_ floatcost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ floatcost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ Sequence[Aicost_ lists Gateway Model Model Target Config Ollama Cache Write Cost List] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ Sequence[Aifactors Gateway Model Model Target Config Ollama Context Window Factor] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions float - The number of dimensions for embedding outputs.
- input_
cost float - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens float - The maximum number of tokens to generate in the response.
- output_
cost float - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ Sequence[Aifactors Gateway Model Model Target Config Ollama Service Tier Factor] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature float
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k float - Limits the number of highest-probability tokens considered during generation.
- top_
p float - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url str - The upstream URL for the model endpoint.
- cache
Read NumberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write NumberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<Property Map>Cost Lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<Property Map>Factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Number - The number of dimensions for embedding outputs.
- input
Cost Number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Number - The maximum number of tokens to generate in the response.
- output
Cost Number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<Property Map>Factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Number - Limits the number of highest-probability tokens considered during generation.
- top
P Number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
AiGatewayModelModelTargetConfigOllamaCacheWriteCostList, AiGatewayModelModelTargetConfigOllamaCacheWriteCostListArgs
AiGatewayModelModelTargetConfigOllamaContextWindowFactor, AiGatewayModelModelTargetConfigOllamaContextWindowFactorArgs
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor double - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor double - Multiplier applied to output pricing above the threshold. Not Null
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor float64 - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor float64 - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor number - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor number - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Double - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Double - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor number - Multiplier applied to output pricing above the threshold. Not Null
- above str
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor float - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor float - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Number - Multiplier applied to output pricing above the threshold. Not Null
AiGatewayModelModelTargetConfigOllamaServiceTierFactor, AiGatewayModelModelTargetConfigOllamaServiceTierFactorArgs
- Factor double
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- Factor float64
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Double
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor float
- Multiplier applied to the whole request for this service tier. Not Null
- tier str
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Number
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
AiGatewayModelModelTargetConfigOpenai, AiGatewayModelModelTargetConfigOpenaiArgs
- Cache
Read doubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write doubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write List<AiCost Lists Gateway Model Model Target Config Openai Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window List<AiFactors Gateway Model Model Target Config Openai Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions double - The number of dimensions for embedding outputs.
- Input
Cost double - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens double - The maximum number of tokens to generate in the response.
- Output
Cost double - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier List<AiFactors Gateway Model Model Target Config Openai Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature double
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K double - Limits the number of highest-probability tokens considered during generation.
- Top
P double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- Cache
Read float64Cost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write float64Cost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write []AiCost Lists Gateway Model Model Target Config Openai Cache Write Cost List - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window []AiFactors Gateway Model Model Target Config Openai Context Window Factor - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions float64 - The number of dimensions for embedding outputs.
- Input
Cost float64 - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens float64 - The maximum number of tokens to generate in the response.
- Output
Cost float64 - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier []AiFactors Gateway Model Model Target Config Openai Service Tier Factor - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature float64
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K float64 - Limits the number of highest-probability tokens considered during generation.
- Top
P float64 - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- cache_
read_ numbercost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ numbercost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ list(object)cost_ lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ list(object)factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions number - The number of dimensions for embedding outputs.
- input_
cost number - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens number - The maximum number of tokens to generate in the response.
- output_
cost number - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ list(object)factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k number - Limits the number of highest-probability tokens considered during generation.
- top_
p number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url string - The upstream URL for the model endpoint.
- cache
Read DoubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write DoubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<AiCost Lists Gateway Model Model Target Config Openai Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<AiFactors Gateway Model Model Target Config Openai Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Double - The number of dimensions for embedding outputs.
- input
Cost Double - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Double - The maximum number of tokens to generate in the response.
- output
Cost Double - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<AiFactors Gateway Model Model Target Config Openai Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Double
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Double - Limits the number of highest-probability tokens considered during generation.
- top
P Double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
- cache
Read numberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write numberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write AiCost Lists Gateway Model Model Target Config Openai Cache Write Cost List[] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window AiFactors Gateway Model Model Target Config Openai Context Window Factor[] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions number - The number of dimensions for embedding outputs.
- input
Cost number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens number - The maximum number of tokens to generate in the response.
- output
Cost number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier AiFactors Gateway Model Model Target Config Openai Service Tier Factor[] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K number - Limits the number of highest-probability tokens considered during generation.
- top
P number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url string - The upstream URL for the model endpoint.
- cache_
read_ floatcost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ floatcost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ Sequence[Aicost_ lists Gateway Model Model Target Config Openai Cache Write Cost List] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ Sequence[Aifactors Gateway Model Model Target Config Openai Context Window Factor] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions float - The number of dimensions for embedding outputs.
- input_
cost float - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens float - The maximum number of tokens to generate in the response.
- output_
cost float - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ Sequence[Aifactors Gateway Model Model Target Config Openai Service Tier Factor] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature float
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k float - Limits the number of highest-probability tokens considered during generation.
- top_
p float - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url str - The upstream URL for the model endpoint.
- cache
Read NumberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write NumberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<Property Map>Cost Lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<Property Map>Factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Number - The number of dimensions for embedding outputs.
- input
Cost Number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Number - The maximum number of tokens to generate in the response.
- output
Cost Number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<Property Map>Factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Number - Limits the number of highest-probability tokens considered during generation.
- top
P Number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
AiGatewayModelModelTargetConfigOpenaiCacheWriteCostList, AiGatewayModelModelTargetConfigOpenaiCacheWriteCostListArgs
AiGatewayModelModelTargetConfigOpenaiContextWindowFactor, AiGatewayModelModelTargetConfigOpenaiContextWindowFactorArgs
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor double - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor double - Multiplier applied to output pricing above the threshold. Not Null
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor float64 - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor float64 - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor number - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor number - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Double - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Double - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor number - Multiplier applied to output pricing above the threshold. Not Null
- above str
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor float - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor float - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Number - Multiplier applied to output pricing above the threshold. Not Null
AiGatewayModelModelTargetConfigOpenaiServiceTierFactor, AiGatewayModelModelTargetConfigOpenaiServiceTierFactorArgs
- Factor double
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- Factor float64
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Double
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor float
- Multiplier applied to the whole request for this service tier. Not Null
- tier str
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Number
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
AiGatewayModelModelTargetConfigSagemaker, AiGatewayModelModelTargetConfigSagemakerArgs
- Aws
Ai
Gateway Model Model Target Config Sagemaker Aws - Cache
Read doubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write doubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write List<AiCost Lists Gateway Model Model Target Config Sagemaker Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window List<AiFactors Gateway Model Model Target Config Sagemaker Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions double - The number of dimensions for embedding outputs.
- Input
Cost double - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens double - The maximum number of tokens to generate in the response.
- Output
Cost double - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier List<AiFactors Gateway Model Model Target Config Sagemaker Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Target
Ai
Gateway Model Model Target Config Sagemaker Target - Temperature double
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K double - Limits the number of highest-probability tokens considered during generation.
- Top
P double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- Aws
Ai
Gateway Model Model Target Config Sagemaker Aws - Cache
Read float64Cost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write float64Cost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write []AiCost Lists Gateway Model Model Target Config Sagemaker Cache Write Cost List - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window []AiFactors Gateway Model Model Target Config Sagemaker Context Window Factor - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions float64 - The number of dimensions for embedding outputs.
- Input
Cost float64 - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens float64 - The maximum number of tokens to generate in the response.
- Output
Cost float64 - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier []AiFactors Gateway Model Model Target Config Sagemaker Service Tier Factor - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Target
Ai
Gateway Model Model Target Config Sagemaker Target - Temperature float64
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K float64 - Limits the number of highest-probability tokens considered during generation.
- Top
P float64 - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- aws object
- cache_
read_ numbercost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ numbercost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ list(object)cost_ lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ list(object)factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions number - The number of dimensions for embedding outputs.
- input_
cost number - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens number - The maximum number of tokens to generate in the response.
- output_
cost number - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ list(object)factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- target object
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k number - Limits the number of highest-probability tokens considered during generation.
- top_
p number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url string - The upstream URL for the model endpoint.
- aws
Ai
Gateway Model Model Target Config Sagemaker Aws - cache
Read DoubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write DoubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<AiCost Lists Gateway Model Model Target Config Sagemaker Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<AiFactors Gateway Model Model Target Config Sagemaker Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Double - The number of dimensions for embedding outputs.
- input
Cost Double - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Double - The maximum number of tokens to generate in the response.
- output
Cost Double - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<AiFactors Gateway Model Model Target Config Sagemaker Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- target
Ai
Gateway Model Model Target Config Sagemaker Target - temperature Double
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Double - Limits the number of highest-probability tokens considered during generation.
- top
P Double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
- aws
Ai
Gateway Model Model Target Config Sagemaker Aws - cache
Read numberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write numberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write AiCost Lists Gateway Model Model Target Config Sagemaker Cache Write Cost List[] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window AiFactors Gateway Model Model Target Config Sagemaker Context Window Factor[] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions number - The number of dimensions for embedding outputs.
- input
Cost number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens number - The maximum number of tokens to generate in the response.
- output
Cost number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier AiFactors Gateway Model Model Target Config Sagemaker Service Tier Factor[] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- target
Ai
Gateway Model Model Target Config Sagemaker Target - temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K number - Limits the number of highest-probability tokens considered during generation.
- top
P number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url string - The upstream URL for the model endpoint.
- aws
Ai
Gateway Model Model Target Config Sagemaker Aws - cache_
read_ floatcost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ floatcost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ Sequence[Aicost_ lists Gateway Model Model Target Config Sagemaker Cache Write Cost List] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ Sequence[Aifactors Gateway Model Model Target Config Sagemaker Context Window Factor] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions float - The number of dimensions for embedding outputs.
- input_
cost float - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens float - The maximum number of tokens to generate in the response.
- output_
cost float - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ Sequence[Aifactors Gateway Model Model Target Config Sagemaker Service Tier Factor] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- target
Ai
Gateway Model Model Target Config Sagemaker Target - temperature float
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k float - Limits the number of highest-probability tokens considered during generation.
- top_
p float - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url str - The upstream URL for the model endpoint.
- aws Property Map
- cache
Read NumberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write NumberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<Property Map>Cost Lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<Property Map>Factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Number - The number of dimensions for embedding outputs.
- input
Cost Number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Number - The maximum number of tokens to generate in the response.
- output
Cost Number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<Property Map>Factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- target Property Map
- temperature Number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Number - Limits the number of highest-probability tokens considered during generation.
- top
P Number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
AiGatewayModelModelTargetConfigSagemakerAws, AiGatewayModelModelTargetConfigSagemakerAwsArgs
- Assume
Role stringArn - Assume a different IAM role after authenticating; mutually required with rolesessionname.
- Region string
- Overrides the AWS_REGION environment variable for SageMaker requests.
- Role
Session stringName - Session identifier for the assumed role; mutually required with assumerolearn.
- Sts
Endpoint stringUrl - Overrides the STS endpoint when assuming a role.
- Assume
Role stringArn - Assume a different IAM role after authenticating; mutually required with rolesessionname.
- Region string
- Overrides the AWS_REGION environment variable for SageMaker requests.
- Role
Session stringName - Session identifier for the assumed role; mutually required with assumerolearn.
- Sts
Endpoint stringUrl - Overrides the STS endpoint when assuming a role.
- assume_
role_ stringarn - Assume a different IAM role after authenticating; mutually required with rolesessionname.
- region string
- Overrides the AWS_REGION environment variable for SageMaker requests.
- role_
session_ stringname - Session identifier for the assumed role; mutually required with assumerolearn.
- sts_
endpoint_ stringurl - Overrides the STS endpoint when assuming a role.
- assume
Role StringArn - Assume a different IAM role after authenticating; mutually required with rolesessionname.
- region String
- Overrides the AWS_REGION environment variable for SageMaker requests.
- role
Session StringName - Session identifier for the assumed role; mutually required with assumerolearn.
- sts
Endpoint StringUrl - Overrides the STS endpoint when assuming a role.
- assume
Role stringArn - Assume a different IAM role after authenticating; mutually required with rolesessionname.
- region string
- Overrides the AWS_REGION environment variable for SageMaker requests.
- role
Session stringName - Session identifier for the assumed role; mutually required with assumerolearn.
- sts
Endpoint stringUrl - Overrides the STS endpoint when assuming a role.
- assume_
role_ strarn - Assume a different IAM role after authenticating; mutually required with rolesessionname.
- region str
- Overrides the AWS_REGION environment variable for SageMaker requests.
- role_
session_ strname - Session identifier for the assumed role; mutually required with assumerolearn.
- sts_
endpoint_ strurl - Overrides the STS endpoint when assuming a role.
- assume
Role StringArn - Assume a different IAM role after authenticating; mutually required with rolesessionname.
- region String
- Overrides the AWS_REGION environment variable for SageMaker requests.
- role
Session StringName - Session identifier for the assumed role; mutually required with assumerolearn.
- sts
Endpoint StringUrl - Overrides the STS endpoint when assuming a role.
AiGatewayModelModelTargetConfigSagemakerCacheWriteCostList, AiGatewayModelModelTargetConfigSagemakerCacheWriteCostListArgs
AiGatewayModelModelTargetConfigSagemakerContextWindowFactor, AiGatewayModelModelTargetConfigSagemakerContextWindowFactorArgs
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor double - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor double - Multiplier applied to output pricing above the threshold. Not Null
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor float64 - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor float64 - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor number - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor number - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Double - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Double - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor number - Multiplier applied to output pricing above the threshold. Not Null
- above str
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor float - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor float - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Number - Multiplier applied to output pricing above the threshold. Not Null
AiGatewayModelModelTargetConfigSagemakerServiceTierFactor, AiGatewayModelModelTargetConfigSagemakerServiceTierFactorArgs
- Factor double
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- Factor float64
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Double
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor float
- Multiplier applied to the whole request for this service tier. Not Null
- tier str
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Number
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
AiGatewayModelModelTargetConfigSagemakerTarget, AiGatewayModelModelTargetConfigSagemakerTargetArgs
- Container
Hostname string - Sets the X-Amzn-SageMaker-Target-Container-Hostname header (multi-container).
- Model string
- Sets the X-Amzn-SageMaker-Target-Model header (multi-model endpoints).
- Variant string
- Sets the X-Amzn-SageMaker-Target-Variant header (A/B variant testing).
- Container
Hostname string - Sets the X-Amzn-SageMaker-Target-Container-Hostname header (multi-container).
- Model string
- Sets the X-Amzn-SageMaker-Target-Model header (multi-model endpoints).
- Variant string
- Sets the X-Amzn-SageMaker-Target-Variant header (A/B variant testing).
- container_
hostname string - Sets the X-Amzn-SageMaker-Target-Container-Hostname header (multi-container).
- model string
- Sets the X-Amzn-SageMaker-Target-Model header (multi-model endpoints).
- variant string
- Sets the X-Amzn-SageMaker-Target-Variant header (A/B variant testing).
- container
Hostname String - Sets the X-Amzn-SageMaker-Target-Container-Hostname header (multi-container).
- model String
- Sets the X-Amzn-SageMaker-Target-Model header (multi-model endpoints).
- variant String
- Sets the X-Amzn-SageMaker-Target-Variant header (A/B variant testing).
- container
Hostname string - Sets the X-Amzn-SageMaker-Target-Container-Hostname header (multi-container).
- model string
- Sets the X-Amzn-SageMaker-Target-Model header (multi-model endpoints).
- variant string
- Sets the X-Amzn-SageMaker-Target-Variant header (A/B variant testing).
- container_
hostname str - Sets the X-Amzn-SageMaker-Target-Container-Hostname header (multi-container).
- model str
- Sets the X-Amzn-SageMaker-Target-Model header (multi-model endpoints).
- variant str
- Sets the X-Amzn-SageMaker-Target-Variant header (A/B variant testing).
- container
Hostname String - Sets the X-Amzn-SageMaker-Target-Container-Hostname header (multi-container).
- model String
- Sets the X-Amzn-SageMaker-Target-Model header (multi-model endpoints).
- variant String
- Sets the X-Amzn-SageMaker-Target-Variant header (A/B variant testing).
AiGatewayModelModelTargetConfigVercel, AiGatewayModelModelTargetConfigVercelArgs
- Cache
Read doubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write doubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write List<AiCost Lists Gateway Model Model Target Config Vercel Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window List<AiFactors Gateway Model Model Target Config Vercel Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions double - The number of dimensions for embedding outputs.
- Input
Cost double - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens double - The maximum number of tokens to generate in the response.
- Output
Cost double - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier List<AiFactors Gateway Model Model Target Config Vercel Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature double
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K double - Limits the number of highest-probability tokens considered during generation.
- Top
P double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- Cache
Read float64Cost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write float64Cost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write []AiCost Lists Gateway Model Model Target Config Vercel Cache Write Cost List - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window []AiFactors Gateway Model Model Target Config Vercel Context Window Factor - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions float64 - The number of dimensions for embedding outputs.
- Input
Cost float64 - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens float64 - The maximum number of tokens to generate in the response.
- Output
Cost float64 - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier []AiFactors Gateway Model Model Target Config Vercel Service Tier Factor - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature float64
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K float64 - Limits the number of highest-probability tokens considered during generation.
- Top
P float64 - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- cache_
read_ numbercost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ numbercost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ list(object)cost_ lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ list(object)factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions number - The number of dimensions for embedding outputs.
- input_
cost number - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens number - The maximum number of tokens to generate in the response.
- output_
cost number - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ list(object)factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k number - Limits the number of highest-probability tokens considered during generation.
- top_
p number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url string - The upstream URL for the model endpoint.
- cache
Read DoubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write DoubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<AiCost Lists Gateway Model Model Target Config Vercel Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<AiFactors Gateway Model Model Target Config Vercel Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Double - The number of dimensions for embedding outputs.
- input
Cost Double - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Double - The maximum number of tokens to generate in the response.
- output
Cost Double - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<AiFactors Gateway Model Model Target Config Vercel Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Double
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Double - Limits the number of highest-probability tokens considered during generation.
- top
P Double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
- cache
Read numberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write numberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write AiCost Lists Gateway Model Model Target Config Vercel Cache Write Cost List[] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window AiFactors Gateway Model Model Target Config Vercel Context Window Factor[] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions number - The number of dimensions for embedding outputs.
- input
Cost number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens number - The maximum number of tokens to generate in the response.
- output
Cost number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier AiFactors Gateway Model Model Target Config Vercel Service Tier Factor[] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K number - Limits the number of highest-probability tokens considered during generation.
- top
P number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url string - The upstream URL for the model endpoint.
- cache_
read_ floatcost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ floatcost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ Sequence[Aicost_ lists Gateway Model Model Target Config Vercel Cache Write Cost List] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ Sequence[Aifactors Gateway Model Model Target Config Vercel Context Window Factor] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions float - The number of dimensions for embedding outputs.
- input_
cost float - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens float - The maximum number of tokens to generate in the response.
- output_
cost float - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ Sequence[Aifactors Gateway Model Model Target Config Vercel Service Tier Factor] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature float
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k float - Limits the number of highest-probability tokens considered during generation.
- top_
p float - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url str - The upstream URL for the model endpoint.
- cache
Read NumberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write NumberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<Property Map>Cost Lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<Property Map>Factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Number - The number of dimensions for embedding outputs.
- input
Cost Number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Number - The maximum number of tokens to generate in the response.
- output
Cost Number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<Property Map>Factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Number - Limits the number of highest-probability tokens considered during generation.
- top
P Number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
AiGatewayModelModelTargetConfigVercelCacheWriteCostList, AiGatewayModelModelTargetConfigVercelCacheWriteCostListArgs
AiGatewayModelModelTargetConfigVercelContextWindowFactor, AiGatewayModelModelTargetConfigVercelContextWindowFactorArgs
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor double - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor double - Multiplier applied to output pricing above the threshold. Not Null
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor float64 - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor float64 - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor number - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor number - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Double - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Double - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor number - Multiplier applied to output pricing above the threshold. Not Null
- above str
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor float - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor float - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Number - Multiplier applied to output pricing above the threshold. Not Null
AiGatewayModelModelTargetConfigVercelServiceTierFactor, AiGatewayModelModelTargetConfigVercelServiceTierFactorArgs
- Factor double
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- Factor float64
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Double
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor float
- Multiplier applied to the whole request for this service tier. Not Null
- tier str
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Number
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
AiGatewayModelModelTargetConfigVllm, AiGatewayModelModelTargetConfigVllmArgs
- Cache
Read doubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write doubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write List<AiCost Lists Gateway Model Model Target Config Vllm Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window List<AiFactors Gateway Model Model Target Config Vllm Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions double - The number of dimensions for embedding outputs.
- Input
Cost double - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens double - The maximum number of tokens to generate in the response.
- Output
Cost double - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier List<AiFactors Gateway Model Model Target Config Vllm Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature double
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K double - Limits the number of highest-probability tokens considered during generation.
- Top
P double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint. Not Null
- Cache
Read float64Cost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write float64Cost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write []AiCost Lists Gateway Model Model Target Config Vllm Cache Write Cost List - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window []AiFactors Gateway Model Model Target Config Vllm Context Window Factor - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions float64 - The number of dimensions for embedding outputs.
- Input
Cost float64 - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens float64 - The maximum number of tokens to generate in the response.
- Output
Cost float64 - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier []AiFactors Gateway Model Model Target Config Vllm Service Tier Factor - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature float64
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K float64 - Limits the number of highest-probability tokens considered during generation.
- Top
P float64 - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint. Not Null
- cache_
read_ numbercost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ numbercost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ list(object)cost_ lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ list(object)factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions number - The number of dimensions for embedding outputs.
- input_
cost number - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens number - The maximum number of tokens to generate in the response.
- output_
cost number - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ list(object)factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k number - Limits the number of highest-probability tokens considered during generation.
- top_
p number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url string - The upstream URL for the model endpoint. Not Null
- cache
Read DoubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write DoubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<AiCost Lists Gateway Model Model Target Config Vllm Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<AiFactors Gateway Model Model Target Config Vllm Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Double - The number of dimensions for embedding outputs.
- input
Cost Double - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Double - The maximum number of tokens to generate in the response.
- output
Cost Double - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<AiFactors Gateway Model Model Target Config Vllm Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Double
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Double - Limits the number of highest-probability tokens considered during generation.
- top
P Double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint. Not Null
- cache
Read numberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write numberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write AiCost Lists Gateway Model Model Target Config Vllm Cache Write Cost List[] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window AiFactors Gateway Model Model Target Config Vllm Context Window Factor[] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions number - The number of dimensions for embedding outputs.
- input
Cost number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens number - The maximum number of tokens to generate in the response.
- output
Cost number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier AiFactors Gateway Model Model Target Config Vllm Service Tier Factor[] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K number - Limits the number of highest-probability tokens considered during generation.
- top
P number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url string - The upstream URL for the model endpoint. Not Null
- cache_
read_ floatcost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ floatcost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ Sequence[Aicost_ lists Gateway Model Model Target Config Vllm Cache Write Cost List] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ Sequence[Aifactors Gateway Model Model Target Config Vllm Context Window Factor] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions float - The number of dimensions for embedding outputs.
- input_
cost float - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens float - The maximum number of tokens to generate in the response.
- output_
cost float - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ Sequence[Aifactors Gateway Model Model Target Config Vllm Service Tier Factor] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature float
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k float - Limits the number of highest-probability tokens considered during generation.
- top_
p float - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url str - The upstream URL for the model endpoint. Not Null
- cache
Read NumberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write NumberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<Property Map>Cost Lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<Property Map>Factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Number - The number of dimensions for embedding outputs.
- input
Cost Number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Number - The maximum number of tokens to generate in the response.
- output
Cost Number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<Property Map>Factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Number - Limits the number of highest-probability tokens considered during generation.
- top
P Number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint. Not Null
AiGatewayModelModelTargetConfigVllmCacheWriteCostList, AiGatewayModelModelTargetConfigVllmCacheWriteCostListArgs
AiGatewayModelModelTargetConfigVllmContextWindowFactor, AiGatewayModelModelTargetConfigVllmContextWindowFactorArgs
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor double - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor double - Multiplier applied to output pricing above the threshold. Not Null
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor float64 - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor float64 - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor number - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor number - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Double - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Double - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor number - Multiplier applied to output pricing above the threshold. Not Null
- above str
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor float - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor float - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Number - Multiplier applied to output pricing above the threshold. Not Null
AiGatewayModelModelTargetConfigVllmServiceTierFactor, AiGatewayModelModelTargetConfigVllmServiceTierFactorArgs
- Factor double
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- Factor float64
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Double
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor float
- Multiplier applied to the whole request for this service tier. Not Null
- tier str
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Number
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
AiGatewayModelModelTargetConfigXai, AiGatewayModelModelTargetConfigXaiArgs
- Cache
Read doubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write doubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write List<AiCost Lists Gateway Model Model Target Config Xai Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window List<AiFactors Gateway Model Model Target Config Xai Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions double - The number of dimensions for embedding outputs.
- Input
Cost double - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens double - The maximum number of tokens to generate in the response.
- Output
Cost double - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier List<AiFactors Gateway Model Model Target Config Xai Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature double
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K double - Limits the number of highest-probability tokens considered during generation.
- Top
P double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- Cache
Read float64Cost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- Cache
Write float64Cost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- Cache
Write []AiCost Lists Gateway Model Model Target Config Xai Cache Write Cost List - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- Context
Window []AiFactors Gateway Model Model Target Config Xai Context Window Factor - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- Embeddings
Dimensions float64 - The number of dimensions for embedding outputs.
- Input
Cost float64 - Cost per 1M input tokens for billing and cost tracking.
- Max
Tokens float64 - The maximum number of tokens to generate in the response.
- Output
Cost float64 - Cost per 1M output tokens for billing and cost tracking.
- Service
Tier []AiFactors Gateway Model Model Target Config Xai Service Tier Factor - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- Temperature float64
- Controls randomness in the model output. Higher values produce more varied responses.
- Top
K float64 - Limits the number of highest-probability tokens considered during generation.
- Top
P float64 - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- Upstream
Url string - The upstream URL for the model endpoint.
- cache_
read_ numbercost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ numbercost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ list(object)cost_ lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ list(object)factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions number - The number of dimensions for embedding outputs.
- input_
cost number - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens number - The maximum number of tokens to generate in the response.
- output_
cost number - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ list(object)factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k number - Limits the number of highest-probability tokens considered during generation.
- top_
p number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url string - The upstream URL for the model endpoint.
- cache
Read DoubleCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write DoubleCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<AiCost Lists Gateway Model Model Target Config Xai Cache Write Cost List> - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<AiFactors Gateway Model Model Target Config Xai Context Window Factor> - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Double - The number of dimensions for embedding outputs.
- input
Cost Double - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Double - The maximum number of tokens to generate in the response.
- output
Cost Double - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<AiFactors Gateway Model Model Target Config Xai Service Tier Factor> - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Double
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Double - Limits the number of highest-probability tokens considered during generation.
- top
P Double - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
- cache
Read numberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write numberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write AiCost Lists Gateway Model Model Target Config Xai Cache Write Cost List[] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window AiFactors Gateway Model Model Target Config Xai Context Window Factor[] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions number - The number of dimensions for embedding outputs.
- input
Cost number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens number - The maximum number of tokens to generate in the response.
- output
Cost number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier AiFactors Gateway Model Model Target Config Xai Service Tier Factor[] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K number - Limits the number of highest-probability tokens considered during generation.
- top
P number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url string - The upstream URL for the model endpoint.
- cache_
read_ floatcost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache_
write_ floatcost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache_
write_ Sequence[Aicost_ lists Gateway Model Model Target Config Xai Cache Write Cost List] - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context_
window_ Sequence[Aifactors Gateway Model Model Target Config Xai Context Window Factor] - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings_
dimensions float - The number of dimensions for embedding outputs.
- input_
cost float - Cost per 1M input tokens for billing and cost tracking.
- max_
tokens float - The maximum number of tokens to generate in the response.
- output_
cost float - Cost per 1M output tokens for billing and cost tracking.
- service_
tier_ Sequence[Aifactors Gateway Model Model Target Config Xai Service Tier Factor] - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature float
- Controls randomness in the model output. Higher values produce more varied responses.
- top_
k float - Limits the number of highest-probability tokens considered during generation.
- top_
p float - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream_
url str - The upstream URL for the model endpoint.
- cache
Read NumberCost - Cost per 1M cache-read (cached) prompt tokens for billing and cost tracking.
- cache
Write NumberCost - Cost per 1M cache-write prompt tokens for billing and cost tracking.
- cache
Write List<Property Map>Cost Lists - Per-cache-TTL cache-write pricing; overrides cachewritecost per TTL. Configure this when the upstream provider charges differently for different cache TTLs.
- context
Window List<Property Map>Factors - Above an input-token threshold, scale input and output pricing by the corresponding factor.
- embeddings
Dimensions Number - The number of dimensions for embedding outputs.
- input
Cost Number - Cost per 1M input tokens for billing and cost tracking.
- max
Tokens Number - The maximum number of tokens to generate in the response.
- output
Cost Number - Cost per 1M output tokens for billing and cost tracking.
- service
Tier List<Property Map>Factors - Multiplier applied to the whole request for a service tier. The default factor is 1.0 when no tier matches.
- temperature Number
- Controls randomness in the model output. Higher values produce more varied responses.
- top
K Number - Limits the number of highest-probability tokens considered during generation.
- top
P Number - Nucleus sampling probability mass. Tokens with cumulative probability up to top_p are considered.
- upstream
Url String - The upstream URL for the model endpoint.
AiGatewayModelModelTargetConfigXaiCacheWriteCostList, AiGatewayModelModelTargetConfigXaiCacheWriteCostListArgs
AiGatewayModelModelTargetConfigXaiContextWindowFactor, AiGatewayModelModelTargetConfigXaiContextWindowFactorArgs
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor double - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor double - Multiplier applied to output pricing above the threshold. Not Null
- Above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- Input
Factor float64 - Multiplier applied to input pricing above the threshold. Not Null
- Output
Factor float64 - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor number - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor number - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Double - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Double - Multiplier applied to output pricing above the threshold. Not Null
- above string
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor number - Multiplier applied to output pricing above the threshold. Not Null
- above str
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input_
factor float - Multiplier applied to input pricing above the threshold. Not Null
- output_
factor float - Multiplier applied to output pricing above the threshold. Not Null
- above String
- Input-token threshold above which the factors apply, e.g. "128k" or "1m". Not Null
- input
Factor Number - Multiplier applied to input pricing above the threshold. Not Null
- output
Factor Number - Multiplier applied to output pricing above the threshold. Not Null
AiGatewayModelModelTargetConfigXaiServiceTierFactor, AiGatewayModelModelTargetConfigXaiServiceTierFactorArgs
- Factor double
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- Factor float64
- Multiplier applied to the whole request for this service tier. Not Null
- Tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Double
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor number
- Multiplier applied to the whole request for this service tier. Not Null
- tier string
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor float
- Multiplier applied to the whole request for this service tier. Not Null
- tier str
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
- factor Number
- Multiplier applied to the whole request for this service tier. Not Null
- tier String
- Matched case-insensitively as a substring of the vendor's reported service tier (e.g. "priority", "flex", "throughput"). When more than one entry matches, the longest (most specific) tier wins; array order does not matter. Not Null
Import
In Terraform v1.5.0 and later, the import block can be used with the id attribute, for example:
terraform
import {
to = konnect_ai_gateway_model.my_konnect_ai_gateway_model
id = jsonencode({
gateway_id = "5f9fd312-a987-4628-b4c5-bb4f4fddd5f7"
id = "5f9fd312-a987-4628-b4c5-bb4f4fddd5f7"
})
}
The pulumi import command can be used, for example:
$ pulumi import konnect:index/aiGatewayModel:AiGatewayModel my_konnect_ai_gateway_model '{"gateway_id": "5f9fd312-a987-4628-b4c5-bb4f4fddd5f7", "id": "5f9fd312-a987-4628-b4c5-bb4f4fddd5f7"}'
To learn more about importing existing cloud resources, see Importing resources.
Package Details
- Repository
- konnect kong/terraform-provider-konnect
- License
- Notes
- This Pulumi package is based on the
konnectTerraform Provider.
Viewing docs for konnect 3.23.0
published on Friday, Sep 18, 2026 by kong
published on Friday, Sep 18, 2026 by kong