databricks.getAiGatewayModelService

Databricks v1.103.0, Aug 11 26

Viewing docs for Databricks v1.103.0
published on Tuesday, Aug 11, 2026 by Pulumi

Schema (JSON)

pulumi/pulumi-databricks

Viewing docs for Databricks v1.103.0
published on Tuesday, Aug 11, 2026 by Pulumi

Schema (JSON)

pulumi/pulumi-databricks

Using getAiGatewayModelService

Two invocation forms are available. The direct form accepts plain arguments and either blocks until the result value is available, or returns a Promise-wrapped result. The output form accepts Input-wrapped arguments and returns an Output-wrapped result.

function getAiGatewayModelService(args: GetAiGatewayModelServiceArgs, opts?: InvokeOptions): Promise<GetAiGatewayModelServiceResult>
function getAiGatewayModelServiceOutput(args: GetAiGatewayModelServiceOutputArgs, opts?: InvokeOptions): Output<GetAiGatewayModelServiceResult>

def get_ai_gateway_model_service(name: Optional[str] = None,
                                 provider_config: Optional[GetAiGatewayModelServiceProviderConfig] = None,
                                 opts: Optional[InvokeOptions] = None) -> GetAiGatewayModelServiceResult
def get_ai_gateway_model_service_output(name: pulumi.Input[Optional[str]] = None,
                                 provider_config: pulumi.Input[Optional[GetAiGatewayModelServiceProviderConfigArgs]] = None,
                                 opts: Optional[InvokeOptions] = None) -> Output[GetAiGatewayModelServiceResult]

func LookupAiGatewayModelService(ctx *Context, args *LookupAiGatewayModelServiceArgs, opts ...InvokeOption) (*LookupAiGatewayModelServiceResult, error)
func LookupAiGatewayModelServiceOutput(ctx *Context, args *LookupAiGatewayModelServiceOutputArgs, opts ...InvokeOption) LookupAiGatewayModelServiceResultOutput

> Note: This function is named LookupAiGatewayModelService in the Go SDK.

public static class GetAiGatewayModelService 
{
    public static Task<GetAiGatewayModelServiceResult> InvokeAsync(GetAiGatewayModelServiceArgs args, InvokeOptions? opts = null)
    public static Output<GetAiGatewayModelServiceResult> Invoke(GetAiGatewayModelServiceInvokeArgs args, InvokeOptions? opts = null)
}

public static CompletableFuture<GetAiGatewayModelServiceResult> getAiGatewayModelService(GetAiGatewayModelServiceArgs args, InvokeOptions options)
public static Output<GetAiGatewayModelServiceResult> getAiGatewayModelService(GetAiGatewayModelServiceArgs args, InvokeOptions options)

fn::invoke:
  function: databricks:index/getAiGatewayModelService:getAiGatewayModelService
  arguments:
    # arguments dictionary

data "databricks_get_ai_gateway_model_service" "name" {
    # arguments
}

The following arguments are supported:

Name string: Resource name of the model service. Format: model-services/{catalog}.{schema}.{model_service}. Each {...} component is capped at 255 characters individually. Server-derived on Create from parent + modelServiceId; required and immutable on Update/Get/Delete
ProviderConfig GetAiGatewayModelServiceProviderConfig: Configure the provider for management through account provider.

Name string: Resource name of the model service. Format: model-services/{catalog}.{schema}.{model_service}. Each {...} component is capped at 255 characters individually. Server-derived on Create from parent + modelServiceId; required and immutable on Update/Get/Delete
ProviderConfig GetAiGatewayModelServiceProviderConfig: Configure the provider for management through account provider.

name string: Resource name of the model service. Format: model-services/{catalog}.{schema}.{model_service}. Each {...} component is capped at 255 characters individually. Server-derived on Create from parent + modelServiceId; required and immutable on Update/Get/Delete
provider_config object: Configure the provider for management through account provider.

name String: Resource name of the model service. Format: model-services/{catalog}.{schema}.{model_service}. Each {...} component is capped at 255 characters individually. Server-derived on Create from parent + modelServiceId; required and immutable on Update/Get/Delete
providerConfig GetAiGatewayModelServiceProviderConfig: Configure the provider for management through account provider.

name string: Resource name of the model service. Format: model-services/{catalog}.{schema}.{model_service}. Each {...} component is capped at 255 characters individually. Server-derived on Create from parent + modelServiceId; required and immutable on Update/Get/Delete
providerConfig GetAiGatewayModelServiceProviderConfig: Configure the provider for management through account provider.

name str: Resource name of the model service. Format: model-services/{catalog}.{schema}.{model_service}. Each {...} component is capped at 255 characters individually. Server-derived on Create from parent + modelServiceId; required and immutable on Update/Get/Delete
provider_config GetAiGatewayModelServiceProviderConfig: Configure the provider for management through account provider.

name String: Resource name of the model service. Format: model-services/{catalog}.{schema}.{model_service}. Each {...} component is capped at 255 characters individually. Server-derived on Create from parent + modelServiceId; required and immutable on Update/Get/Delete
providerConfig Property Map: Configure the provider for management through account provider.

getAiGatewayModelService Result

The following output properties are available:

BrowseOnly bool: (boolean) - Whether the caller sees only metadata available through the BROWSE privilege
Comment string: (string) - User-provided description
Config GetAiGatewayModelServiceConfig: (ModelServiceConfig) - Operational configuration: destinations, routing, rate limits, inference table. Required on CreateModelService; on UpdateModelService it is required only when config (or a config.* subpath) appears in updateMask
CreateTime string: (string) - When the model service was created
CreatedBy string: (string) - Creator identity
EffectiveOwner string: (string) - The resolved owner of the ModelService. Falls back to the caller's identity when owner is not explicitly set on creation
Etag string: (string) - Optimistic concurrency control token. Server-generated from the entity's state and returned on every read. To use it as an if-match precondition on a mutation, echo the last-read value back via the dedicated etag field on the Update / Delete request; the server rejects the mutation if the stored etag differs
MetastoreId string: (string) - Metastore hosting the model service
Name string: (string) - User-facing label for this destination, used in routing references
Owner string: (string) - The owner of the model service. Write-only; read owner via effective_owner
SupportedApiTypes List<string>: (list of string) - Unified API types this endpoint supports (e.g. "chat", "embeddings", "completions"). Derived from the destinations' backing models / providers at read time
UpdateTime string: (string) - When the model service was last modified
UpdatedBy string: (string) - Identity of the last updater
ProviderConfig GetAiGatewayModelServiceProviderConfig

BrowseOnly bool: (boolean) - Whether the caller sees only metadata available through the BROWSE privilege
Comment string: (string) - User-provided description
Config GetAiGatewayModelServiceConfig: (ModelServiceConfig) - Operational configuration: destinations, routing, rate limits, inference table. Required on CreateModelService; on UpdateModelService it is required only when config (or a config.* subpath) appears in updateMask
CreateTime string: (string) - When the model service was created
CreatedBy string: (string) - Creator identity
EffectiveOwner string: (string) - The resolved owner of the ModelService. Falls back to the caller's identity when owner is not explicitly set on creation
Etag string: (string) - Optimistic concurrency control token. Server-generated from the entity's state and returned on every read. To use it as an if-match precondition on a mutation, echo the last-read value back via the dedicated etag field on the Update / Delete request; the server rejects the mutation if the stored etag differs
MetastoreId string: (string) - Metastore hosting the model service
Name string: (string) - User-facing label for this destination, used in routing references
Owner string: (string) - The owner of the model service. Write-only; read owner via effective_owner
SupportedApiTypes []string: (list of string) - Unified API types this endpoint supports (e.g. "chat", "embeddings", "completions"). Derived from the destinations' backing models / providers at read time
UpdateTime string: (string) - When the model service was last modified
UpdatedBy string: (string) - Identity of the last updater
ProviderConfig GetAiGatewayModelServiceProviderConfig

browse_only bool: (boolean) - Whether the caller sees only metadata available through the BROWSE privilege
comment string: (string) - User-provided description
config object: (ModelServiceConfig) - Operational configuration: destinations, routing, rate limits, inference table. Required on CreateModelService; on UpdateModelService it is required only when config (or a config.* subpath) appears in updateMask
create_time string: (string) - When the model service was created
created_by string: (string) - Creator identity
effective_owner string: (string) - The resolved owner of the ModelService. Falls back to the caller's identity when owner is not explicitly set on creation
etag string: (string) - Optimistic concurrency control token. Server-generated from the entity's state and returned on every read. To use it as an if-match precondition on a mutation, echo the last-read value back via the dedicated etag field on the Update / Delete request; the server rejects the mutation if the stored etag differs
metastore_id string: (string) - Metastore hosting the model service
name string: (string) - User-facing label for this destination, used in routing references
owner string: (string) - The owner of the model service. Write-only; read owner via effective_owner
supported_api_types list(string): (list of string) - Unified API types this endpoint supports (e.g. "chat", "embeddings", "completions"). Derived from the destinations' backing models / providers at read time
update_time string: (string) - When the model service was last modified
updated_by string: (string) - Identity of the last updater
provider_config object

browseOnly Boolean: (boolean) - Whether the caller sees only metadata available through the BROWSE privilege
comment String: (string) - User-provided description
config GetAiGatewayModelServiceConfig: (ModelServiceConfig) - Operational configuration: destinations, routing, rate limits, inference table. Required on CreateModelService; on UpdateModelService it is required only when config (or a config.* subpath) appears in updateMask
createTime String: (string) - When the model service was created
createdBy String: (string) - Creator identity
effectiveOwner String: (string) - The resolved owner of the ModelService. Falls back to the caller's identity when owner is not explicitly set on creation
etag String: (string) - Optimistic concurrency control token. Server-generated from the entity's state and returned on every read. To use it as an if-match precondition on a mutation, echo the last-read value back via the dedicated etag field on the Update / Delete request; the server rejects the mutation if the stored etag differs
metastoreId String: (string) - Metastore hosting the model service
name String: (string) - User-facing label for this destination, used in routing references
owner String: (string) - The owner of the model service. Write-only; read owner via effective_owner
supportedApiTypes List<String>: (list of string) - Unified API types this endpoint supports (e.g. "chat", "embeddings", "completions"). Derived from the destinations' backing models / providers at read time
updateTime String: (string) - When the model service was last modified
updatedBy String: (string) - Identity of the last updater
providerConfig GetAiGatewayModelServiceProviderConfig

browseOnly boolean: (boolean) - Whether the caller sees only metadata available through the BROWSE privilege
comment string: (string) - User-provided description
config GetAiGatewayModelServiceConfig: (ModelServiceConfig) - Operational configuration: destinations, routing, rate limits, inference table. Required on CreateModelService; on UpdateModelService it is required only when config (or a config.* subpath) appears in updateMask
createTime string: (string) - When the model service was created
createdBy string: (string) - Creator identity
effectiveOwner string: (string) - The resolved owner of the ModelService. Falls back to the caller's identity when owner is not explicitly set on creation
etag string: (string) - Optimistic concurrency control token. Server-generated from the entity's state and returned on every read. To use it as an if-match precondition on a mutation, echo the last-read value back via the dedicated etag field on the Update / Delete request; the server rejects the mutation if the stored etag differs
metastoreId string: (string) - Metastore hosting the model service
name string: (string) - User-facing label for this destination, used in routing references
owner string: (string) - The owner of the model service. Write-only; read owner via effective_owner
supportedApiTypes string[]: (list of string) - Unified API types this endpoint supports (e.g. "chat", "embeddings", "completions"). Derived from the destinations' backing models / providers at read time
updateTime string: (string) - When the model service was last modified
updatedBy string: (string) - Identity of the last updater
providerConfig GetAiGatewayModelServiceProviderConfig

browse_only bool: (boolean) - Whether the caller sees only metadata available through the BROWSE privilege
comment str: (string) - User-provided description
config GetAiGatewayModelServiceConfig: (ModelServiceConfig) - Operational configuration: destinations, routing, rate limits, inference table. Required on CreateModelService; on UpdateModelService it is required only when config (or a config.* subpath) appears in updateMask
create_time str: (string) - When the model service was created
created_by str: (string) - Creator identity
effective_owner str: (string) - The resolved owner of the ModelService. Falls back to the caller's identity when owner is not explicitly set on creation
etag str: (string) - Optimistic concurrency control token. Server-generated from the entity's state and returned on every read. To use it as an if-match precondition on a mutation, echo the last-read value back via the dedicated etag field on the Update / Delete request; the server rejects the mutation if the stored etag differs
metastore_id str: (string) - Metastore hosting the model service
name str: (string) - User-facing label for this destination, used in routing references
owner str: (string) - The owner of the model service. Write-only; read owner via effective_owner
supported_api_types Sequence[str]: (list of string) - Unified API types this endpoint supports (e.g. "chat", "embeddings", "completions"). Derived from the destinations' backing models / providers at read time
update_time str: (string) - When the model service was last modified
updated_by str: (string) - Identity of the last updater
provider_config GetAiGatewayModelServiceProviderConfig

browseOnly Boolean: (boolean) - Whether the caller sees only metadata available through the BROWSE privilege
comment String: (string) - User-provided description
config Property Map: (ModelServiceConfig) - Operational configuration: destinations, routing, rate limits, inference table. Required on CreateModelService; on UpdateModelService it is required only when config (or a config.* subpath) appears in updateMask
createTime String: (string) - When the model service was created
createdBy String: (string) - Creator identity
effectiveOwner String: (string) - The resolved owner of the ModelService. Falls back to the caller's identity when owner is not explicitly set on creation
etag String: (string) - Optimistic concurrency control token. Server-generated from the entity's state and returned on every read. To use it as an if-match precondition on a mutation, echo the last-read value back via the dedicated etag field on the Update / Delete request; the server rejects the mutation if the stored etag differs
metastoreId String: (string) - Metastore hosting the model service
name String: (string) - User-facing label for this destination, used in routing references
owner String: (string) - The owner of the model service. Write-only; read owner via effective_owner
supportedApiTypes List<String>: (list of string) - Unified API types this endpoint supports (e.g. "chat", "embeddings", "completions"). Derived from the destinations' backing models / providers at read time
updateTime String: (string) - When the model service was last modified
updatedBy String: (string) - Identity of the last updater
providerConfig Property Map

Supporting Types

GetAiGatewayModelServiceConfig

InferenceTable GetAiGatewayModelServiceConfigInferenceTable: (InferenceTableConfig) - Inference table config for payload logging
RateLimits List<GetAiGatewayModelServiceConfigRateLimit>: (list of RateLimit) - Rate limits applied to requests routed through this model service
Routing GetAiGatewayModelServiceConfigRouting: (ModelServiceConfigRoutingConfig) - Routing configuration: destinations, routing strategy, and fallback

InferenceTable GetAiGatewayModelServiceConfigInferenceTable: (InferenceTableConfig) - Inference table config for payload logging
RateLimits []GetAiGatewayModelServiceConfigRateLimit: (list of RateLimit) - Rate limits applied to requests routed through this model service
Routing GetAiGatewayModelServiceConfigRouting: (ModelServiceConfigRoutingConfig) - Routing configuration: destinations, routing strategy, and fallback

inference_table object: (InferenceTableConfig) - Inference table config for payload logging
rate_limits list(object): (list of RateLimit) - Rate limits applied to requests routed through this model service
routing object: (ModelServiceConfigRoutingConfig) - Routing configuration: destinations, routing strategy, and fallback

inferenceTable GetAiGatewayModelServiceConfigInferenceTable: (InferenceTableConfig) - Inference table config for payload logging
rateLimits List<GetAiGatewayModelServiceConfigRateLimit>: (list of RateLimit) - Rate limits applied to requests routed through this model service
routing GetAiGatewayModelServiceConfigRouting: (ModelServiceConfigRoutingConfig) - Routing configuration: destinations, routing strategy, and fallback

inferenceTable GetAiGatewayModelServiceConfigInferenceTable: (InferenceTableConfig) - Inference table config for payload logging
rateLimits GetAiGatewayModelServiceConfigRateLimit[]: (list of RateLimit) - Rate limits applied to requests routed through this model service
routing GetAiGatewayModelServiceConfigRouting: (ModelServiceConfigRoutingConfig) - Routing configuration: destinations, routing strategy, and fallback

inference_table GetAiGatewayModelServiceConfigInferenceTable: (InferenceTableConfig) - Inference table config for payload logging
rate_limits Sequence[GetAiGatewayModelServiceConfigRateLimit]: (list of RateLimit) - Rate limits applied to requests routed through this model service
routing GetAiGatewayModelServiceConfigRouting: (ModelServiceConfigRoutingConfig) - Routing configuration: destinations, routing strategy, and fallback

inferenceTable Property Map: (InferenceTableConfig) - Inference table config for payload logging
rateLimits List<Property Map>: (list of RateLimit) - Rate limits applied to requests routed through this model service
routing Property Map: (ModelServiceConfigRoutingConfig) - Routing configuration: destinations, routing strategy, and fallback

GetAiGatewayModelServiceConfigInferenceTable

IsDeleted bool: (boolean) - True when the destination's backing UC entity (MODEL for foundation-model destinations, MODEL_PROVIDER_SERVICE for external destinations) has been deleted but the destination row still references it. The dangling destination is surfaced (not silently dropped) so callers can see the broken routing. Inference traffic through this destination fails closed (BAD_REQUEST / FAILED_PRECONDITION)
Parent string: (string) - Parent UC schema where the inference table is created. Format: schemas/{catalog}.{schema}. Set at create time and immutable thereafter; changing it on an existing service is rejected
Table string: (string) - Resolved UC table for payload logs. Format: tables/{catalog}.{schema}.{table}
Disabled bool: (boolean) - Indicates whether payload logging is disabled (opt-out). Unset means that payload logging is active (the on-by-default state coincides with the proto zero-value, so the server never fills this field for a client that leaves it unset). Set disabled = true to pause runtime logging while keeping the sub-message attached (preserving parent and tableNamePrefix for a later flip back to active). parent remains required either way
TableNamePrefix string: (string) - Prefix for the inference-table's UC-registered name. The actual leaf name UC stores is <table_name_prefix>_payload; the _payload suffix is appended automatically. To find the actual UC table after Create, read the table field on the response. Defaults to <model_service_name>_payload when unset. Set at create time and immutable thereafter; changing it on an existing service is rejected

IsDeleted bool: (boolean) - True when the destination's backing UC entity (MODEL for foundation-model destinations, MODEL_PROVIDER_SERVICE for external destinations) has been deleted but the destination row still references it. The dangling destination is surfaced (not silently dropped) so callers can see the broken routing. Inference traffic through this destination fails closed (BAD_REQUEST / FAILED_PRECONDITION)
Parent string: (string) - Parent UC schema where the inference table is created. Format: schemas/{catalog}.{schema}. Set at create time and immutable thereafter; changing it on an existing service is rejected
Table string: (string) - Resolved UC table for payload logs. Format: tables/{catalog}.{schema}.{table}
Disabled bool: (boolean) - Indicates whether payload logging is disabled (opt-out). Unset means that payload logging is active (the on-by-default state coincides with the proto zero-value, so the server never fills this field for a client that leaves it unset). Set disabled = true to pause runtime logging while keeping the sub-message attached (preserving parent and tableNamePrefix for a later flip back to active). parent remains required either way
TableNamePrefix string: (string) - Prefix for the inference-table's UC-registered name. The actual leaf name UC stores is <table_name_prefix>_payload; the _payload suffix is appended automatically. To find the actual UC table after Create, read the table field on the response. Defaults to <model_service_name>_payload when unset. Set at create time and immutable thereafter; changing it on an existing service is rejected

is_deleted bool: (boolean) - True when the destination's backing UC entity (MODEL for foundation-model destinations, MODEL_PROVIDER_SERVICE for external destinations) has been deleted but the destination row still references it. The dangling destination is surfaced (not silently dropped) so callers can see the broken routing. Inference traffic through this destination fails closed (BAD_REQUEST / FAILED_PRECONDITION)
parent string: (string) - Parent UC schema where the inference table is created. Format: schemas/{catalog}.{schema}. Set at create time and immutable thereafter; changing it on an existing service is rejected
table string: (string) - Resolved UC table for payload logs. Format: tables/{catalog}.{schema}.{table}
disabled bool: (boolean) - Indicates whether payload logging is disabled (opt-out). Unset means that payload logging is active (the on-by-default state coincides with the proto zero-value, so the server never fills this field for a client that leaves it unset). Set disabled = true to pause runtime logging while keeping the sub-message attached (preserving parent and tableNamePrefix for a later flip back to active). parent remains required either way
table_name_prefix string: (string) - Prefix for the inference-table's UC-registered name. The actual leaf name UC stores is <table_name_prefix>_payload; the _payload suffix is appended automatically. To find the actual UC table after Create, read the table field on the response. Defaults to <model_service_name>_payload when unset. Set at create time and immutable thereafter; changing it on an existing service is rejected

isDeleted Boolean: (boolean) - True when the destination's backing UC entity (MODEL for foundation-model destinations, MODEL_PROVIDER_SERVICE for external destinations) has been deleted but the destination row still references it. The dangling destination is surfaced (not silently dropped) so callers can see the broken routing. Inference traffic through this destination fails closed (BAD_REQUEST / FAILED_PRECONDITION)
parent String: (string) - Parent UC schema where the inference table is created. Format: schemas/{catalog}.{schema}. Set at create time and immutable thereafter; changing it on an existing service is rejected
table String: (string) - Resolved UC table for payload logs. Format: tables/{catalog}.{schema}.{table}
disabled Boolean: (boolean) - Indicates whether payload logging is disabled (opt-out). Unset means that payload logging is active (the on-by-default state coincides with the proto zero-value, so the server never fills this field for a client that leaves it unset). Set disabled = true to pause runtime logging while keeping the sub-message attached (preserving parent and tableNamePrefix for a later flip back to active). parent remains required either way
tableNamePrefix String: (string) - Prefix for the inference-table's UC-registered name. The actual leaf name UC stores is <table_name_prefix>_payload; the _payload suffix is appended automatically. To find the actual UC table after Create, read the table field on the response. Defaults to <model_service_name>_payload when unset. Set at create time and immutable thereafter; changing it on an existing service is rejected

isDeleted boolean: (boolean) - True when the destination's backing UC entity (MODEL for foundation-model destinations, MODEL_PROVIDER_SERVICE for external destinations) has been deleted but the destination row still references it. The dangling destination is surfaced (not silently dropped) so callers can see the broken routing. Inference traffic through this destination fails closed (BAD_REQUEST / FAILED_PRECONDITION)
parent string: (string) - Parent UC schema where the inference table is created. Format: schemas/{catalog}.{schema}. Set at create time and immutable thereafter; changing it on an existing service is rejected
table string: (string) - Resolved UC table for payload logs. Format: tables/{catalog}.{schema}.{table}
disabled boolean: (boolean) - Indicates whether payload logging is disabled (opt-out). Unset means that payload logging is active (the on-by-default state coincides with the proto zero-value, so the server never fills this field for a client that leaves it unset). Set disabled = true to pause runtime logging while keeping the sub-message attached (preserving parent and tableNamePrefix for a later flip back to active). parent remains required either way
tableNamePrefix string: (string) - Prefix for the inference-table's UC-registered name. The actual leaf name UC stores is <table_name_prefix>_payload; the _payload suffix is appended automatically. To find the actual UC table after Create, read the table field on the response. Defaults to <model_service_name>_payload when unset. Set at create time and immutable thereafter; changing it on an existing service is rejected

is_deleted bool: (boolean) - True when the destination's backing UC entity (MODEL for foundation-model destinations, MODEL_PROVIDER_SERVICE for external destinations) has been deleted but the destination row still references it. The dangling destination is surfaced (not silently dropped) so callers can see the broken routing. Inference traffic through this destination fails closed (BAD_REQUEST / FAILED_PRECONDITION)
parent str: (string) - Parent UC schema where the inference table is created. Format: schemas/{catalog}.{schema}. Set at create time and immutable thereafter; changing it on an existing service is rejected
table str: (string) - Resolved UC table for payload logs. Format: tables/{catalog}.{schema}.{table}
disabled bool: (boolean) - Indicates whether payload logging is disabled (opt-out). Unset means that payload logging is active (the on-by-default state coincides with the proto zero-value, so the server never fills this field for a client that leaves it unset). Set disabled = true to pause runtime logging while keeping the sub-message attached (preserving parent and tableNamePrefix for a later flip back to active). parent remains required either way
table_name_prefix str: (string) - Prefix for the inference-table's UC-registered name. The actual leaf name UC stores is <table_name_prefix>_payload; the _payload suffix is appended automatically. To find the actual UC table after Create, read the table field on the response. Defaults to <model_service_name>_payload when unset. Set at create time and immutable thereafter; changing it on an existing service is rejected

isDeleted Boolean: (boolean) - True when the destination's backing UC entity (MODEL for foundation-model destinations, MODEL_PROVIDER_SERVICE for external destinations) has been deleted but the destination row still references it. The dangling destination is surfaced (not silently dropped) so callers can see the broken routing. Inference traffic through this destination fails closed (BAD_REQUEST / FAILED_PRECONDITION)
parent String: (string) - Parent UC schema where the inference table is created. Format: schemas/{catalog}.{schema}. Set at create time and immutable thereafter; changing it on an existing service is rejected
table String: (string) - Resolved UC table for payload logs. Format: tables/{catalog}.{schema}.{table}
disabled Boolean: (boolean) - Indicates whether payload logging is disabled (opt-out). Unset means that payload logging is active (the on-by-default state coincides with the proto zero-value, so the server never fills this field for a client that leaves it unset). Set disabled = true to pause runtime logging while keeping the sub-message attached (preserving parent and tableNamePrefix for a later flip back to active). parent remains required either way
tableNamePrefix String: (string) - Prefix for the inference-table's UC-registered name. The actual leaf name UC stores is <table_name_prefix>_payload; the _payload suffix is appended automatically. To find the actual UC table after Create, read the table field on the response. Defaults to <model_service_name>_payload when unset. Set at create time and immutable thereafter; changing it on an existing service is rejected

GetAiGatewayModelServiceConfigRateLimit

Key string: (string) - Scope key. Determines whether principal is required. Possible values are: RATE_LIMIT_KEY_REQUEST_TAG, RATE_LIMIT_KEY_SERVICE, RATE_LIMIT_KEY_SERVICE_PRINCIPAL, RATE_LIMIT_KEY_USER, RATE_LIMIT_KEY_USER_DEFAULT, RATE_LIMIT_KEY_USER_GROUP
RenewalPeriod string: (string) - Renewal period. Possible values are: RATE_LIMIT_RENEWAL_PERIOD_HOUR, RATE_LIMIT_RENEWAL_PERIOD_MINUTE
Principal string: (string) - Principal this limit applies to: user email, group name, or service principal application ID. Required unless key is RATE_LIMIT_KEY_SERVICE, RATE_LIMIT_KEY_USER_DEFAULT, or RATE_LIMIT_KEY_REQUEST_TAG (which must not set a principal)
RequestTagKey string: (string) - Request tag key this limit applies to. Required when key is RATE_LIMIT_KEY_REQUEST_TAG, forbidden otherwise
RequestTagValue string: (string) - Request tag value this limit applies to. Only valid when key is RATE_LIMIT_KEY_REQUEST_TAG. Leave unset to apply the limit to every value of requestTagKey (an any-value default); a set value is a specific override for that value
Requests int: (integer) - Max requests allowed within a renewal period. Leave unset for no request limit
Tokens int: (integer) - Max tokens allowed within a renewal period. Leave unset for no token limit

Key string: (string) - Scope key. Determines whether principal is required. Possible values are: RATE_LIMIT_KEY_REQUEST_TAG, RATE_LIMIT_KEY_SERVICE, RATE_LIMIT_KEY_SERVICE_PRINCIPAL, RATE_LIMIT_KEY_USER, RATE_LIMIT_KEY_USER_DEFAULT, RATE_LIMIT_KEY_USER_GROUP
RenewalPeriod string: (string) - Renewal period. Possible values are: RATE_LIMIT_RENEWAL_PERIOD_HOUR, RATE_LIMIT_RENEWAL_PERIOD_MINUTE
Principal string: (string) - Principal this limit applies to: user email, group name, or service principal application ID. Required unless key is RATE_LIMIT_KEY_SERVICE, RATE_LIMIT_KEY_USER_DEFAULT, or RATE_LIMIT_KEY_REQUEST_TAG (which must not set a principal)
RequestTagKey string: (string) - Request tag key this limit applies to. Required when key is RATE_LIMIT_KEY_REQUEST_TAG, forbidden otherwise
RequestTagValue string: (string) - Request tag value this limit applies to. Only valid when key is RATE_LIMIT_KEY_REQUEST_TAG. Leave unset to apply the limit to every value of requestTagKey (an any-value default); a set value is a specific override for that value
Requests int: (integer) - Max requests allowed within a renewal period. Leave unset for no request limit
Tokens int: (integer) - Max tokens allowed within a renewal period. Leave unset for no token limit

key string: (string) - Scope key. Determines whether principal is required. Possible values are: RATE_LIMIT_KEY_REQUEST_TAG, RATE_LIMIT_KEY_SERVICE, RATE_LIMIT_KEY_SERVICE_PRINCIPAL, RATE_LIMIT_KEY_USER, RATE_LIMIT_KEY_USER_DEFAULT, RATE_LIMIT_KEY_USER_GROUP
renewal_period string: (string) - Renewal period. Possible values are: RATE_LIMIT_RENEWAL_PERIOD_HOUR, RATE_LIMIT_RENEWAL_PERIOD_MINUTE
principal string: (string) - Principal this limit applies to: user email, group name, or service principal application ID. Required unless key is RATE_LIMIT_KEY_SERVICE, RATE_LIMIT_KEY_USER_DEFAULT, or RATE_LIMIT_KEY_REQUEST_TAG (which must not set a principal)
request_tag_key string: (string) - Request tag key this limit applies to. Required when key is RATE_LIMIT_KEY_REQUEST_TAG, forbidden otherwise
request_tag_value string: (string) - Request tag value this limit applies to. Only valid when key is RATE_LIMIT_KEY_REQUEST_TAG. Leave unset to apply the limit to every value of requestTagKey (an any-value default); a set value is a specific override for that value
requests number: (integer) - Max requests allowed within a renewal period. Leave unset for no request limit
tokens number: (integer) - Max tokens allowed within a renewal period. Leave unset for no token limit

key String: (string) - Scope key. Determines whether principal is required. Possible values are: RATE_LIMIT_KEY_REQUEST_TAG, RATE_LIMIT_KEY_SERVICE, RATE_LIMIT_KEY_SERVICE_PRINCIPAL, RATE_LIMIT_KEY_USER, RATE_LIMIT_KEY_USER_DEFAULT, RATE_LIMIT_KEY_USER_GROUP
renewalPeriod String: (string) - Renewal period. Possible values are: RATE_LIMIT_RENEWAL_PERIOD_HOUR, RATE_LIMIT_RENEWAL_PERIOD_MINUTE
principal String: (string) - Principal this limit applies to: user email, group name, or service principal application ID. Required unless key is RATE_LIMIT_KEY_SERVICE, RATE_LIMIT_KEY_USER_DEFAULT, or RATE_LIMIT_KEY_REQUEST_TAG (which must not set a principal)
requestTagKey String: (string) - Request tag key this limit applies to. Required when key is RATE_LIMIT_KEY_REQUEST_TAG, forbidden otherwise
requestTagValue String: (string) - Request tag value this limit applies to. Only valid when key is RATE_LIMIT_KEY_REQUEST_TAG. Leave unset to apply the limit to every value of requestTagKey (an any-value default); a set value is a specific override for that value
requests Integer: (integer) - Max requests allowed within a renewal period. Leave unset for no request limit
tokens Integer: (integer) - Max tokens allowed within a renewal period. Leave unset for no token limit

key string: (string) - Scope key. Determines whether principal is required. Possible values are: RATE_LIMIT_KEY_REQUEST_TAG, RATE_LIMIT_KEY_SERVICE, RATE_LIMIT_KEY_SERVICE_PRINCIPAL, RATE_LIMIT_KEY_USER, RATE_LIMIT_KEY_USER_DEFAULT, RATE_LIMIT_KEY_USER_GROUP
renewalPeriod string: (string) - Renewal period. Possible values are: RATE_LIMIT_RENEWAL_PERIOD_HOUR, RATE_LIMIT_RENEWAL_PERIOD_MINUTE
principal string: (string) - Principal this limit applies to: user email, group name, or service principal application ID. Required unless key is RATE_LIMIT_KEY_SERVICE, RATE_LIMIT_KEY_USER_DEFAULT, or RATE_LIMIT_KEY_REQUEST_TAG (which must not set a principal)
requestTagKey string: (string) - Request tag key this limit applies to. Required when key is RATE_LIMIT_KEY_REQUEST_TAG, forbidden otherwise
requestTagValue string: (string) - Request tag value this limit applies to. Only valid when key is RATE_LIMIT_KEY_REQUEST_TAG. Leave unset to apply the limit to every value of requestTagKey (an any-value default); a set value is a specific override for that value
requests number: (integer) - Max requests allowed within a renewal period. Leave unset for no request limit
tokens number: (integer) - Max tokens allowed within a renewal period. Leave unset for no token limit

key str: (string) - Scope key. Determines whether principal is required. Possible values are: RATE_LIMIT_KEY_REQUEST_TAG, RATE_LIMIT_KEY_SERVICE, RATE_LIMIT_KEY_SERVICE_PRINCIPAL, RATE_LIMIT_KEY_USER, RATE_LIMIT_KEY_USER_DEFAULT, RATE_LIMIT_KEY_USER_GROUP
renewal_period str: (string) - Renewal period. Possible values are: RATE_LIMIT_RENEWAL_PERIOD_HOUR, RATE_LIMIT_RENEWAL_PERIOD_MINUTE
principal str: (string) - Principal this limit applies to: user email, group name, or service principal application ID. Required unless key is RATE_LIMIT_KEY_SERVICE, RATE_LIMIT_KEY_USER_DEFAULT, or RATE_LIMIT_KEY_REQUEST_TAG (which must not set a principal)
request_tag_key str: (string) - Request tag key this limit applies to. Required when key is RATE_LIMIT_KEY_REQUEST_TAG, forbidden otherwise
request_tag_value str: (string) - Request tag value this limit applies to. Only valid when key is RATE_LIMIT_KEY_REQUEST_TAG. Leave unset to apply the limit to every value of requestTagKey (an any-value default); a set value is a specific override for that value
requests int: (integer) - Max requests allowed within a renewal period. Leave unset for no request limit
tokens int: (integer) - Max tokens allowed within a renewal period. Leave unset for no token limit

key String: (string) - Scope key. Determines whether principal is required. Possible values are: RATE_LIMIT_KEY_REQUEST_TAG, RATE_LIMIT_KEY_SERVICE, RATE_LIMIT_KEY_SERVICE_PRINCIPAL, RATE_LIMIT_KEY_USER, RATE_LIMIT_KEY_USER_DEFAULT, RATE_LIMIT_KEY_USER_GROUP
renewalPeriod String: (string) - Renewal period. Possible values are: RATE_LIMIT_RENEWAL_PERIOD_HOUR, RATE_LIMIT_RENEWAL_PERIOD_MINUTE
principal String: (string) - Principal this limit applies to: user email, group name, or service principal application ID. Required unless key is RATE_LIMIT_KEY_SERVICE, RATE_LIMIT_KEY_USER_DEFAULT, or RATE_LIMIT_KEY_REQUEST_TAG (which must not set a principal)
requestTagKey String: (string) - Request tag key this limit applies to. Required when key is RATE_LIMIT_KEY_REQUEST_TAG, forbidden otherwise
requestTagValue String: (string) - Request tag value this limit applies to. Only valid when key is RATE_LIMIT_KEY_REQUEST_TAG. Leave unset to apply the limit to every value of requestTagKey (an any-value default); a set value is a specific override for that value
requests Number: (integer) - Max requests allowed within a renewal period. Leave unset for no request limit
tokens Number: (integer) - Max tokens allowed within a renewal period. Leave unset for no token limit

GetAiGatewayModelServiceConfigRouting

Destinations List<GetAiGatewayModelServiceConfigRoutingDestination>: (list of ModelServiceConfigDestinationConfig) - Primary routing destinations. At most 10 are allowed. At least one is required on CreateModelService; on UpdateModelService it is required only when config.routing (or a config.routing.* subpath) appears in updateMask
Fallback GetAiGatewayModelServiceConfigRoutingFallback: (ModelServiceConfigFallbackConfig) - Fallback routing config, applied after primary destinations fail
FirstTokenTimeout string: (string) - Timeout for the first token of a streaming response. If a destination does not return its first token within this duration, AI Gateway aborts the attempt and fails over to the next destination. Applies to streaming requests only. Leave unset for no first-token timeout
TrafficSplitting GetAiGatewayModelServiceConfigRoutingTrafficSplitting: (ModelServiceConfigRoutingConfigTrafficSplitting) - Marker message selecting request-based traffic splitting. Traffic is distributed according to each destination's trafficPercentage value; no configuration lives on this message itself

Destinations []GetAiGatewayModelServiceConfigRoutingDestination: (list of ModelServiceConfigDestinationConfig) - Primary routing destinations. At most 10 are allowed. At least one is required on CreateModelService; on UpdateModelService it is required only when config.routing (or a config.routing.* subpath) appears in updateMask
Fallback GetAiGatewayModelServiceConfigRoutingFallback: (ModelServiceConfigFallbackConfig) - Fallback routing config, applied after primary destinations fail
FirstTokenTimeout string: (string) - Timeout for the first token of a streaming response. If a destination does not return its first token within this duration, AI Gateway aborts the attempt and fails over to the next destination. Applies to streaming requests only. Leave unset for no first-token timeout
TrafficSplitting GetAiGatewayModelServiceConfigRoutingTrafficSplitting: (ModelServiceConfigRoutingConfigTrafficSplitting) - Marker message selecting request-based traffic splitting. Traffic is distributed according to each destination's trafficPercentage value; no configuration lives on this message itself

destinations list(object): (list of ModelServiceConfigDestinationConfig) - Primary routing destinations. At most 10 are allowed. At least one is required on CreateModelService; on UpdateModelService it is required only when config.routing (or a config.routing.* subpath) appears in updateMask
fallback object: (ModelServiceConfigFallbackConfig) - Fallback routing config, applied after primary destinations fail
first_token_timeout string: (string) - Timeout for the first token of a streaming response. If a destination does not return its first token within this duration, AI Gateway aborts the attempt and fails over to the next destination. Applies to streaming requests only. Leave unset for no first-token timeout
traffic_splitting object: (ModelServiceConfigRoutingConfigTrafficSplitting) - Marker message selecting request-based traffic splitting. Traffic is distributed according to each destination's trafficPercentage value; no configuration lives on this message itself

destinations List<GetAiGatewayModelServiceConfigRoutingDestination>: (list of ModelServiceConfigDestinationConfig) - Primary routing destinations. At most 10 are allowed. At least one is required on CreateModelService; on UpdateModelService it is required only when config.routing (or a config.routing.* subpath) appears in updateMask
fallback GetAiGatewayModelServiceConfigRoutingFallback: (ModelServiceConfigFallbackConfig) - Fallback routing config, applied after primary destinations fail
firstTokenTimeout String: (string) - Timeout for the first token of a streaming response. If a destination does not return its first token within this duration, AI Gateway aborts the attempt and fails over to the next destination. Applies to streaming requests only. Leave unset for no first-token timeout
trafficSplitting GetAiGatewayModelServiceConfigRoutingTrafficSplitting: (ModelServiceConfigRoutingConfigTrafficSplitting) - Marker message selecting request-based traffic splitting. Traffic is distributed according to each destination's trafficPercentage value; no configuration lives on this message itself

destinations GetAiGatewayModelServiceConfigRoutingDestination[]: (list of ModelServiceConfigDestinationConfig) - Primary routing destinations. At most 10 are allowed. At least one is required on CreateModelService; on UpdateModelService it is required only when config.routing (or a config.routing.* subpath) appears in updateMask
fallback GetAiGatewayModelServiceConfigRoutingFallback: (ModelServiceConfigFallbackConfig) - Fallback routing config, applied after primary destinations fail
firstTokenTimeout string: (string) - Timeout for the first token of a streaming response. If a destination does not return its first token within this duration, AI Gateway aborts the attempt and fails over to the next destination. Applies to streaming requests only. Leave unset for no first-token timeout
trafficSplitting GetAiGatewayModelServiceConfigRoutingTrafficSplitting: (ModelServiceConfigRoutingConfigTrafficSplitting) - Marker message selecting request-based traffic splitting. Traffic is distributed according to each destination's trafficPercentage value; no configuration lives on this message itself

destinations Sequence[GetAiGatewayModelServiceConfigRoutingDestination]: (list of ModelServiceConfigDestinationConfig) - Primary routing destinations. At most 10 are allowed. At least one is required on CreateModelService; on UpdateModelService it is required only when config.routing (or a config.routing.* subpath) appears in updateMask
fallback GetAiGatewayModelServiceConfigRoutingFallback: (ModelServiceConfigFallbackConfig) - Fallback routing config, applied after primary destinations fail
first_token_timeout str: (string) - Timeout for the first token of a streaming response. If a destination does not return its first token within this duration, AI Gateway aborts the attempt and fails over to the next destination. Applies to streaming requests only. Leave unset for no first-token timeout
traffic_splitting GetAiGatewayModelServiceConfigRoutingTrafficSplitting: (ModelServiceConfigRoutingConfigTrafficSplitting) - Marker message selecting request-based traffic splitting. Traffic is distributed according to each destination's trafficPercentage value; no configuration lives on this message itself

destinations List<Property Map>: (list of ModelServiceConfigDestinationConfig) - Primary routing destinations. At most 10 are allowed. At least one is required on CreateModelService; on UpdateModelService it is required only when config.routing (or a config.routing.* subpath) appears in updateMask
fallback Property Map: (ModelServiceConfigFallbackConfig) - Fallback routing config, applied after primary destinations fail
firstTokenTimeout String: (string) - Timeout for the first token of a streaming response. If a destination does not return its first token within this duration, AI Gateway aborts the attempt and fails over to the next destination. Applies to streaming requests only. Leave unset for no first-token timeout
trafficSplitting Property Map: (ModelServiceConfigRoutingConfigTrafficSplitting) - Marker message selecting request-based traffic splitting. Traffic is distributed according to each destination's trafficPercentage value; no configuration lives on this message itself

GetAiGatewayModelServiceConfigRoutingDestination

DestinationType string: (string) - Backing-model category. Determines which oneof variant is populated. Possible values are: DESTINATION_TYPE_EXTERNAL_FOUNDATION_MODEL, DESTINATION_TYPE_PAY_PER_TOKEN_FOUNDATION_MODEL, DESTINATION_TYPE_PROVISIONED_THROUGHPUT_FOUNDATION_MODEL
IsDeleted bool: (boolean) - True when the destination's backing UC entity (MODEL for foundation-model destinations, MODEL_PROVIDER_SERVICE for external destinations) has been deleted but the destination row still references it. The dangling destination is surfaced (not silently dropped) so callers can see the broken routing. Inference traffic through this destination fails closed (BAD_REQUEST / FAILED_PRECONDITION)
Name string: Resource name of the model service. Format: model-services/{catalog}.{schema}.{model_service}. Each {...} component is capped at 255 characters individually. Server-derived on Create from parent + modelServiceId; required and immutable on Update/Get/Delete
ExternalModelConfig GetAiGatewayModelServiceConfigRoutingDestinationExternalModelConfig: (ModelServiceConfigExternalModelConfig)
PayPerTokenConfig GetAiGatewayModelServiceConfigRoutingDestinationPayPerTokenConfig: (ModelServiceConfigPayPerTokenConfig)
ProvisionedThroughputConfig GetAiGatewayModelServiceConfigRoutingDestinationProvisionedThroughputConfig: (ModelServiceConfigProvisionedThroughputConfig)
TrafficPercentage int: (integer) - Share of traffic sent to this destination, 0-100. Optional on fallback destinations; see FallbackConfig

DestinationType string: (string) - Backing-model category. Determines which oneof variant is populated. Possible values are: DESTINATION_TYPE_EXTERNAL_FOUNDATION_MODEL, DESTINATION_TYPE_PAY_PER_TOKEN_FOUNDATION_MODEL, DESTINATION_TYPE_PROVISIONED_THROUGHPUT_FOUNDATION_MODEL
IsDeleted bool: (boolean) - True when the destination's backing UC entity (MODEL for foundation-model destinations, MODEL_PROVIDER_SERVICE for external destinations) has been deleted but the destination row still references it. The dangling destination is surfaced (not silently dropped) so callers can see the broken routing. Inference traffic through this destination fails closed (BAD_REQUEST / FAILED_PRECONDITION)
Name string: Resource name of the model service. Format: model-services/{catalog}.{schema}.{model_service}. Each {...} component is capped at 255 characters individually. Server-derived on Create from parent + modelServiceId; required and immutable on Update/Get/Delete
ExternalModelConfig GetAiGatewayModelServiceConfigRoutingDestinationExternalModelConfig: (ModelServiceConfigExternalModelConfig)
PayPerTokenConfig GetAiGatewayModelServiceConfigRoutingDestinationPayPerTokenConfig: (ModelServiceConfigPayPerTokenConfig)
ProvisionedThroughputConfig GetAiGatewayModelServiceConfigRoutingDestinationProvisionedThroughputConfig: (ModelServiceConfigProvisionedThroughputConfig)
TrafficPercentage int: (integer) - Share of traffic sent to this destination, 0-100. Optional on fallback destinations; see FallbackConfig

destination_type string: (string) - Backing-model category. Determines which oneof variant is populated. Possible values are: DESTINATION_TYPE_EXTERNAL_FOUNDATION_MODEL, DESTINATION_TYPE_PAY_PER_TOKEN_FOUNDATION_MODEL, DESTINATION_TYPE_PROVISIONED_THROUGHPUT_FOUNDATION_MODEL
is_deleted bool: (boolean) - True when the destination's backing UC entity (MODEL for foundation-model destinations, MODEL_PROVIDER_SERVICE for external destinations) has been deleted but the destination row still references it. The dangling destination is surfaced (not silently dropped) so callers can see the broken routing. Inference traffic through this destination fails closed (BAD_REQUEST / FAILED_PRECONDITION)
name string: Resource name of the model service. Format: model-services/{catalog}.{schema}.{model_service}. Each {...} component is capped at 255 characters individually. Server-derived on Create from parent + modelServiceId; required and immutable on Update/Get/Delete
external_model_config object: (ModelServiceConfigExternalModelConfig)
pay_per_token_config object: (ModelServiceConfigPayPerTokenConfig)
provisioned_throughput_config object: (ModelServiceConfigProvisionedThroughputConfig)
traffic_percentage number: (integer) - Share of traffic sent to this destination, 0-100. Optional on fallback destinations; see FallbackConfig

destinationType String: (string) - Backing-model category. Determines which oneof variant is populated. Possible values are: DESTINATION_TYPE_EXTERNAL_FOUNDATION_MODEL, DESTINATION_TYPE_PAY_PER_TOKEN_FOUNDATION_MODEL, DESTINATION_TYPE_PROVISIONED_THROUGHPUT_FOUNDATION_MODEL
isDeleted Boolean: (boolean) - True when the destination's backing UC entity (MODEL for foundation-model destinations, MODEL_PROVIDER_SERVICE for external destinations) has been deleted but the destination row still references it. The dangling destination is surfaced (not silently dropped) so callers can see the broken routing. Inference traffic through this destination fails closed (BAD_REQUEST / FAILED_PRECONDITION)
name String: Resource name of the model service. Format: model-services/{catalog}.{schema}.{model_service}. Each {...} component is capped at 255 characters individually. Server-derived on Create from parent + modelServiceId; required and immutable on Update/Get/Delete
externalModelConfig GetAiGatewayModelServiceConfigRoutingDestinationExternalModelConfig: (ModelServiceConfigExternalModelConfig)
payPerTokenConfig GetAiGatewayModelServiceConfigRoutingDestinationPayPerTokenConfig: (ModelServiceConfigPayPerTokenConfig)
provisionedThroughputConfig GetAiGatewayModelServiceConfigRoutingDestinationProvisionedThroughputConfig: (ModelServiceConfigProvisionedThroughputConfig)
trafficPercentage Integer: (integer) - Share of traffic sent to this destination, 0-100. Optional on fallback destinations; see FallbackConfig

destinationType string: (string) - Backing-model category. Determines which oneof variant is populated. Possible values are: DESTINATION_TYPE_EXTERNAL_FOUNDATION_MODEL, DESTINATION_TYPE_PAY_PER_TOKEN_FOUNDATION_MODEL, DESTINATION_TYPE_PROVISIONED_THROUGHPUT_FOUNDATION_MODEL
isDeleted boolean: (boolean) - True when the destination's backing UC entity (MODEL for foundation-model destinations, MODEL_PROVIDER_SERVICE for external destinations) has been deleted but the destination row still references it. The dangling destination is surfaced (not silently dropped) so callers can see the broken routing. Inference traffic through this destination fails closed (BAD_REQUEST / FAILED_PRECONDITION)
name string: Resource name of the model service. Format: model-services/{catalog}.{schema}.{model_service}. Each {...} component is capped at 255 characters individually. Server-derived on Create from parent + modelServiceId; required and immutable on Update/Get/Delete
externalModelConfig GetAiGatewayModelServiceConfigRoutingDestinationExternalModelConfig: (ModelServiceConfigExternalModelConfig)
payPerTokenConfig GetAiGatewayModelServiceConfigRoutingDestinationPayPerTokenConfig: (ModelServiceConfigPayPerTokenConfig)
provisionedThroughputConfig GetAiGatewayModelServiceConfigRoutingDestinationProvisionedThroughputConfig: (ModelServiceConfigProvisionedThroughputConfig)
trafficPercentage number: (integer) - Share of traffic sent to this destination, 0-100. Optional on fallback destinations; see FallbackConfig

destination_type str: (string) - Backing-model category. Determines which oneof variant is populated. Possible values are: DESTINATION_TYPE_EXTERNAL_FOUNDATION_MODEL, DESTINATION_TYPE_PAY_PER_TOKEN_FOUNDATION_MODEL, DESTINATION_TYPE_PROVISIONED_THROUGHPUT_FOUNDATION_MODEL
is_deleted bool: (boolean) - True when the destination's backing UC entity (MODEL for foundation-model destinations, MODEL_PROVIDER_SERVICE for external destinations) has been deleted but the destination row still references it. The dangling destination is surfaced (not silently dropped) so callers can see the broken routing. Inference traffic through this destination fails closed (BAD_REQUEST / FAILED_PRECONDITION)
name str: Resource name of the model service. Format: model-services/{catalog}.{schema}.{model_service}. Each {...} component is capped at 255 characters individually. Server-derived on Create from parent + modelServiceId; required and immutable on Update/Get/Delete
external_model_config GetAiGatewayModelServiceConfigRoutingDestinationExternalModelConfig: (ModelServiceConfigExternalModelConfig)
pay_per_token_config GetAiGatewayModelServiceConfigRoutingDestinationPayPerTokenConfig: (ModelServiceConfigPayPerTokenConfig)
provisioned_throughput_config GetAiGatewayModelServiceConfigRoutingDestinationProvisionedThroughputConfig: (ModelServiceConfigProvisionedThroughputConfig)
traffic_percentage int: (integer) - Share of traffic sent to this destination, 0-100. Optional on fallback destinations; see FallbackConfig

destinationType String: (string) - Backing-model category. Determines which oneof variant is populated. Possible values are: DESTINATION_TYPE_EXTERNAL_FOUNDATION_MODEL, DESTINATION_TYPE_PAY_PER_TOKEN_FOUNDATION_MODEL, DESTINATION_TYPE_PROVISIONED_THROUGHPUT_FOUNDATION_MODEL
isDeleted Boolean: (boolean) - True when the destination's backing UC entity (MODEL for foundation-model destinations, MODEL_PROVIDER_SERVICE for external destinations) has been deleted but the destination row still references it. The dangling destination is surfaced (not silently dropped) so callers can see the broken routing. Inference traffic through this destination fails closed (BAD_REQUEST / FAILED_PRECONDITION)
name String: Resource name of the model service. Format: model-services/{catalog}.{schema}.{model_service}. Each {...} component is capped at 255 characters individually. Server-derived on Create from parent + modelServiceId; required and immutable on Update/Get/Delete
externalModelConfig Property Map: (ModelServiceConfigExternalModelConfig)
payPerTokenConfig Property Map: (ModelServiceConfigPayPerTokenConfig)
provisionedThroughputConfig Property Map: (ModelServiceConfigProvisionedThroughputConfig)
trafficPercentage Number: (integer) - Share of traffic sent to this destination, 0-100. Optional on fallback destinations; see FallbackConfig

GetAiGatewayModelServiceConfigRoutingDestinationExternalModelConfig

ModelProviderService string: (string) - Resource name of the governed ModelProviderService that owns provider auth and provider-specific configuration. The referenced ModelProviderService also carries the provider type, so this message does not surface it directly. Format: model-provider-services/{catalog}.{schema}.{model_provider_service}. Each {...} component is capped at 255 characters individually
Target GetAiGatewayModelServiceConfigRoutingDestinationExternalModelConfigTarget: (ModelProviderServiceConfigModelTargetConfig) - Routing target for the destination: the provider-side model selected from the referenced ModelProviderService's targets catalog, plus the unified API types the platform should translate to/from at request time

ModelProviderService string: (string) - Resource name of the governed ModelProviderService that owns provider auth and provider-specific configuration. The referenced ModelProviderService also carries the provider type, so this message does not surface it directly. Format: model-provider-services/{catalog}.{schema}.{model_provider_service}. Each {...} component is capped at 255 characters individually
Target GetAiGatewayModelServiceConfigRoutingDestinationExternalModelConfigTarget: (ModelProviderServiceConfigModelTargetConfig) - Routing target for the destination: the provider-side model selected from the referenced ModelProviderService's targets catalog, plus the unified API types the platform should translate to/from at request time

model_provider_service string: (string) - Resource name of the governed ModelProviderService that owns provider auth and provider-specific configuration. The referenced ModelProviderService also carries the provider type, so this message does not surface it directly. Format: model-provider-services/{catalog}.{schema}.{model_provider_service}. Each {...} component is capped at 255 characters individually
target object: (ModelProviderServiceConfigModelTargetConfig) - Routing target for the destination: the provider-side model selected from the referenced ModelProviderService's targets catalog, plus the unified API types the platform should translate to/from at request time

modelProviderService String: (string) - Resource name of the governed ModelProviderService that owns provider auth and provider-specific configuration. The referenced ModelProviderService also carries the provider type, so this message does not surface it directly. Format: model-provider-services/{catalog}.{schema}.{model_provider_service}. Each {...} component is capped at 255 characters individually
target GetAiGatewayModelServiceConfigRoutingDestinationExternalModelConfigTarget: (ModelProviderServiceConfigModelTargetConfig) - Routing target for the destination: the provider-side model selected from the referenced ModelProviderService's targets catalog, plus the unified API types the platform should translate to/from at request time

modelProviderService string: (string) - Resource name of the governed ModelProviderService that owns provider auth and provider-specific configuration. The referenced ModelProviderService also carries the provider type, so this message does not surface it directly. Format: model-provider-services/{catalog}.{schema}.{model_provider_service}. Each {...} component is capped at 255 characters individually
target GetAiGatewayModelServiceConfigRoutingDestinationExternalModelConfigTarget: (ModelProviderServiceConfigModelTargetConfig) - Routing target for the destination: the provider-side model selected from the referenced ModelProviderService's targets catalog, plus the unified API types the platform should translate to/from at request time

model_provider_service str: (string) - Resource name of the governed ModelProviderService that owns provider auth and provider-specific configuration. The referenced ModelProviderService also carries the provider type, so this message does not surface it directly. Format: model-provider-services/{catalog}.{schema}.{model_provider_service}. Each {...} component is capped at 255 characters individually
target GetAiGatewayModelServiceConfigRoutingDestinationExternalModelConfigTarget: (ModelProviderServiceConfigModelTargetConfig) - Routing target for the destination: the provider-side model selected from the referenced ModelProviderService's targets catalog, plus the unified API types the platform should translate to/from at request time

modelProviderService String: (string) - Resource name of the governed ModelProviderService that owns provider auth and provider-specific configuration. The referenced ModelProviderService also carries the provider type, so this message does not surface it directly. Format: model-provider-services/{catalog}.{schema}.{model_provider_service}. Each {...} component is capped at 255 characters individually
target Property Map: (ModelProviderServiceConfigModelTargetConfig) - Routing target for the destination: the provider-side model selected from the referenced ModelProviderService's targets catalog, plus the unified API types the platform should translate to/from at request time

GetAiGatewayModelServiceConfigRoutingDestinationExternalModelConfigTarget

Model string: (string) - UC model FQN of the model served by the backing endpoint (e.g., system.ai.databricks-claude-opus-4-6). Resolved from Model Serving at Create/Update time
NativeApiTypes List<string>: (list of string) - Provider-native API types the model supports (e.g. "openai/v1/chat/completions"). Used by the platform for request/response translation from the unified API type. At most 64 entries of at most 256 characters each; the list is persisted into the destination binding's bounded storage envelope

Model string: (string) - UC model FQN of the model served by the backing endpoint (e.g., system.ai.databricks-claude-opus-4-6). Resolved from Model Serving at Create/Update time
NativeApiTypes []string: (list of string) - Provider-native API types the model supports (e.g. "openai/v1/chat/completions"). Used by the platform for request/response translation from the unified API type. At most 64 entries of at most 256 characters each; the list is persisted into the destination binding's bounded storage envelope

model string: (string) - UC model FQN of the model served by the backing endpoint (e.g., system.ai.databricks-claude-opus-4-6). Resolved from Model Serving at Create/Update time
native_api_types list(string): (list of string) - Provider-native API types the model supports (e.g. "openai/v1/chat/completions"). Used by the platform for request/response translation from the unified API type. At most 64 entries of at most 256 characters each; the list is persisted into the destination binding's bounded storage envelope

model String: (string) - UC model FQN of the model served by the backing endpoint (e.g., system.ai.databricks-claude-opus-4-6). Resolved from Model Serving at Create/Update time
nativeApiTypes List<String>: (list of string) - Provider-native API types the model supports (e.g. "openai/v1/chat/completions"). Used by the platform for request/response translation from the unified API type. At most 64 entries of at most 256 characters each; the list is persisted into the destination binding's bounded storage envelope

model string: (string) - UC model FQN of the model served by the backing endpoint (e.g., system.ai.databricks-claude-opus-4-6). Resolved from Model Serving at Create/Update time
nativeApiTypes string[]: (list of string) - Provider-native API types the model supports (e.g. "openai/v1/chat/completions"). Used by the platform for request/response translation from the unified API type. At most 64 entries of at most 256 characters each; the list is persisted into the destination binding's bounded storage envelope

model str: (string) - UC model FQN of the model served by the backing endpoint (e.g., system.ai.databricks-claude-opus-4-6). Resolved from Model Serving at Create/Update time
native_api_types Sequence[str]: (list of string) - Provider-native API types the model supports (e.g. "openai/v1/chat/completions"). Used by the platform for request/response translation from the unified API type. At most 64 entries of at most 256 characters each; the list is persisted into the destination binding's bounded storage envelope

model String: (string) - UC model FQN of the model served by the backing endpoint (e.g., system.ai.databricks-claude-opus-4-6). Resolved from Model Serving at Create/Update time
nativeApiTypes List<String>: (list of string) - Provider-native API types the model supports (e.g. "openai/v1/chat/completions"). Used by the platform for request/response translation from the unified API type. At most 64 entries of at most 256 characters each; the list is persisted into the destination binding's bounded storage envelope

GetAiGatewayModelServiceConfigRoutingDestinationPayPerTokenConfig

Model string: (string) - UC model FQN of the model served by the backing endpoint (e.g., system.ai.databricks-claude-opus-4-6). Resolved from Model Serving at Create/Update time

Model string: (string) - UC model FQN of the model served by the backing endpoint (e.g., system.ai.databricks-claude-opus-4-6). Resolved from Model Serving at Create/Update time

model string: (string) - UC model FQN of the model served by the backing endpoint (e.g., system.ai.databricks-claude-opus-4-6). Resolved from Model Serving at Create/Update time

model String: (string) - UC model FQN of the model served by the backing endpoint (e.g., system.ai.databricks-claude-opus-4-6). Resolved from Model Serving at Create/Update time

model string: (string) - UC model FQN of the model served by the backing endpoint (e.g., system.ai.databricks-claude-opus-4-6). Resolved from Model Serving at Create/Update time

model str: (string) - UC model FQN of the model served by the backing endpoint (e.g., system.ai.databricks-claude-opus-4-6). Resolved from Model Serving at Create/Update time

model String: (string) - UC model FQN of the model served by the backing endpoint (e.g., system.ai.databricks-claude-opus-4-6). Resolved from Model Serving at Create/Update time

GetAiGatewayModelServiceConfigRoutingDestinationProvisionedThroughputConfig

Model string: (string) - UC model FQN of the model served by the backing endpoint (e.g., system.ai.databricks-claude-opus-4-6). Resolved from Model Serving at Create/Update time
ModelServingEndpoint string: (string) - Name of the backing Model Serving endpoint serving the provisioned- throughput foundation model, as the AIP-122 typed resource name serving-endpoints/{name}. The same UC model can be served on multiple Model Serving endpoints (different throughput / region / config); the caller picks which one this destination routes to. The endpoint must exist at create time

Model string: (string) - UC model FQN of the model served by the backing endpoint (e.g., system.ai.databricks-claude-opus-4-6). Resolved from Model Serving at Create/Update time
ModelServingEndpoint string: (string) - Name of the backing Model Serving endpoint serving the provisioned- throughput foundation model, as the AIP-122 typed resource name serving-endpoints/{name}. The same UC model can be served on multiple Model Serving endpoints (different throughput / region / config); the caller picks which one this destination routes to. The endpoint must exist at create time

model string: (string) - UC model FQN of the model served by the backing endpoint (e.g., system.ai.databricks-claude-opus-4-6). Resolved from Model Serving at Create/Update time
model_serving_endpoint string: (string) - Name of the backing Model Serving endpoint serving the provisioned- throughput foundation model, as the AIP-122 typed resource name serving-endpoints/{name}. The same UC model can be served on multiple Model Serving endpoints (different throughput / region / config); the caller picks which one this destination routes to. The endpoint must exist at create time

model String: (string) - UC model FQN of the model served by the backing endpoint (e.g., system.ai.databricks-claude-opus-4-6). Resolved from Model Serving at Create/Update time
modelServingEndpoint String: (string) - Name of the backing Model Serving endpoint serving the provisioned- throughput foundation model, as the AIP-122 typed resource name serving-endpoints/{name}. The same UC model can be served on multiple Model Serving endpoints (different throughput / region / config); the caller picks which one this destination routes to. The endpoint must exist at create time

model string: (string) - UC model FQN of the model served by the backing endpoint (e.g., system.ai.databricks-claude-opus-4-6). Resolved from Model Serving at Create/Update time
modelServingEndpoint string: (string) - Name of the backing Model Serving endpoint serving the provisioned- throughput foundation model, as the AIP-122 typed resource name serving-endpoints/{name}. The same UC model can be served on multiple Model Serving endpoints (different throughput / region / config); the caller picks which one this destination routes to. The endpoint must exist at create time

model str: (string) - UC model FQN of the model served by the backing endpoint (e.g., system.ai.databricks-claude-opus-4-6). Resolved from Model Serving at Create/Update time
model_serving_endpoint str: (string) - Name of the backing Model Serving endpoint serving the provisioned- throughput foundation model, as the AIP-122 typed resource name serving-endpoints/{name}. The same UC model can be served on multiple Model Serving endpoints (different throughput / region / config); the caller picks which one this destination routes to. The endpoint must exist at create time

model String: (string) - UC model FQN of the model served by the backing endpoint (e.g., system.ai.databricks-claude-opus-4-6). Resolved from Model Serving at Create/Update time
modelServingEndpoint String: (string) - Name of the backing Model Serving endpoint serving the provisioned- throughput foundation model, as the AIP-122 typed resource name serving-endpoints/{name}. The same UC model can be served on multiple Model Serving endpoints (different throughput / region / config); the caller picks which one this destination routes to. The endpoint must exist at create time

GetAiGatewayModelServiceConfigRoutingFallback

Destinations List<GetAiGatewayModelServiceConfigRoutingFallbackDestination>: (list of ModelServiceConfigDestinationConfig) - Primary routing destinations. At most 10 are allowed. At least one is required on CreateModelService; on UpdateModelService it is required only when config.routing (or a config.routing.* subpath) appears in updateMask

Destinations []GetAiGatewayModelServiceConfigRoutingFallbackDestination: (list of ModelServiceConfigDestinationConfig) - Primary routing destinations. At most 10 are allowed. At least one is required on CreateModelService; on UpdateModelService it is required only when config.routing (or a config.routing.* subpath) appears in updateMask

destinations list(object): (list of ModelServiceConfigDestinationConfig) - Primary routing destinations. At most 10 are allowed. At least one is required on CreateModelService; on UpdateModelService it is required only when config.routing (or a config.routing.* subpath) appears in updateMask

destinations List<GetAiGatewayModelServiceConfigRoutingFallbackDestination>: (list of ModelServiceConfigDestinationConfig) - Primary routing destinations. At most 10 are allowed. At least one is required on CreateModelService; on UpdateModelService it is required only when config.routing (or a config.routing.* subpath) appears in updateMask

destinations GetAiGatewayModelServiceConfigRoutingFallbackDestination[]: (list of ModelServiceConfigDestinationConfig) - Primary routing destinations. At most 10 are allowed. At least one is required on CreateModelService; on UpdateModelService it is required only when config.routing (or a config.routing.* subpath) appears in updateMask

destinations Sequence[GetAiGatewayModelServiceConfigRoutingFallbackDestination]: (list of ModelServiceConfigDestinationConfig) - Primary routing destinations. At most 10 are allowed. At least one is required on CreateModelService; on UpdateModelService it is required only when config.routing (or a config.routing.* subpath) appears in updateMask

destinations List<Property Map>: (list of ModelServiceConfigDestinationConfig) - Primary routing destinations. At most 10 are allowed. At least one is required on CreateModelService; on UpdateModelService it is required only when config.routing (or a config.routing.* subpath) appears in updateMask

GetAiGatewayModelServiceConfigRoutingFallbackDestination

DestinationType string: (string) - Backing-model category. Determines which oneof variant is populated. Possible values are: DESTINATION_TYPE_EXTERNAL_FOUNDATION_MODEL, DESTINATION_TYPE_PAY_PER_TOKEN_FOUNDATION_MODEL, DESTINATION_TYPE_PROVISIONED_THROUGHPUT_FOUNDATION_MODEL
IsDeleted bool: (boolean) - True when the destination's backing UC entity (MODEL for foundation-model destinations, MODEL_PROVIDER_SERVICE for external destinations) has been deleted but the destination row still references it. The dangling destination is surfaced (not silently dropped) so callers can see the broken routing. Inference traffic through this destination fails closed (BAD_REQUEST / FAILED_PRECONDITION)
Name string: Resource name of the model service. Format: model-services/{catalog}.{schema}.{model_service}. Each {...} component is capped at 255 characters individually. Server-derived on Create from parent + modelServiceId; required and immutable on Update/Get/Delete
ExternalModelConfig GetAiGatewayModelServiceConfigRoutingFallbackDestinationExternalModelConfig: (ModelServiceConfigExternalModelConfig)
PayPerTokenConfig GetAiGatewayModelServiceConfigRoutingFallbackDestinationPayPerTokenConfig: (ModelServiceConfigPayPerTokenConfig)
ProvisionedThroughputConfig GetAiGatewayModelServiceConfigRoutingFallbackDestinationProvisionedThroughputConfig: (ModelServiceConfigProvisionedThroughputConfig)
TrafficPercentage int: (integer) - Share of traffic sent to this destination, 0-100. Optional on fallback destinations; see FallbackConfig

DestinationType string: (string) - Backing-model category. Determines which oneof variant is populated. Possible values are: DESTINATION_TYPE_EXTERNAL_FOUNDATION_MODEL, DESTINATION_TYPE_PAY_PER_TOKEN_FOUNDATION_MODEL, DESTINATION_TYPE_PROVISIONED_THROUGHPUT_FOUNDATION_MODEL
IsDeleted bool: (boolean) - True when the destination's backing UC entity (MODEL for foundation-model destinations, MODEL_PROVIDER_SERVICE for external destinations) has been deleted but the destination row still references it. The dangling destination is surfaced (not silently dropped) so callers can see the broken routing. Inference traffic through this destination fails closed (BAD_REQUEST / FAILED_PRECONDITION)
Name string: Resource name of the model service. Format: model-services/{catalog}.{schema}.{model_service}. Each {...} component is capped at 255 characters individually. Server-derived on Create from parent + modelServiceId; required and immutable on Update/Get/Delete
ExternalModelConfig GetAiGatewayModelServiceConfigRoutingFallbackDestinationExternalModelConfig: (ModelServiceConfigExternalModelConfig)
PayPerTokenConfig GetAiGatewayModelServiceConfigRoutingFallbackDestinationPayPerTokenConfig: (ModelServiceConfigPayPerTokenConfig)
ProvisionedThroughputConfig GetAiGatewayModelServiceConfigRoutingFallbackDestinationProvisionedThroughputConfig: (ModelServiceConfigProvisionedThroughputConfig)
TrafficPercentage int: (integer) - Share of traffic sent to this destination, 0-100. Optional on fallback destinations; see FallbackConfig

destination_type string: (string) - Backing-model category. Determines which oneof variant is populated. Possible values are: DESTINATION_TYPE_EXTERNAL_FOUNDATION_MODEL, DESTINATION_TYPE_PAY_PER_TOKEN_FOUNDATION_MODEL, DESTINATION_TYPE_PROVISIONED_THROUGHPUT_FOUNDATION_MODEL
is_deleted bool: (boolean) - True when the destination's backing UC entity (MODEL for foundation-model destinations, MODEL_PROVIDER_SERVICE for external destinations) has been deleted but the destination row still references it. The dangling destination is surfaced (not silently dropped) so callers can see the broken routing. Inference traffic through this destination fails closed (BAD_REQUEST / FAILED_PRECONDITION)
name string: Resource name of the model service. Format: model-services/{catalog}.{schema}.{model_service}. Each {...} component is capped at 255 characters individually. Server-derived on Create from parent + modelServiceId; required and immutable on Update/Get/Delete
external_model_config object: (ModelServiceConfigExternalModelConfig)
pay_per_token_config object: (ModelServiceConfigPayPerTokenConfig)
provisioned_throughput_config object: (ModelServiceConfigProvisionedThroughputConfig)
traffic_percentage number: (integer) - Share of traffic sent to this destination, 0-100. Optional on fallback destinations; see FallbackConfig

destinationType String: (string) - Backing-model category. Determines which oneof variant is populated. Possible values are: DESTINATION_TYPE_EXTERNAL_FOUNDATION_MODEL, DESTINATION_TYPE_PAY_PER_TOKEN_FOUNDATION_MODEL, DESTINATION_TYPE_PROVISIONED_THROUGHPUT_FOUNDATION_MODEL
isDeleted Boolean: (boolean) - True when the destination's backing UC entity (MODEL for foundation-model destinations, MODEL_PROVIDER_SERVICE for external destinations) has been deleted but the destination row still references it. The dangling destination is surfaced (not silently dropped) so callers can see the broken routing. Inference traffic through this destination fails closed (BAD_REQUEST / FAILED_PRECONDITION)
name String: Resource name of the model service. Format: model-services/{catalog}.{schema}.{model_service}. Each {...} component is capped at 255 characters individually. Server-derived on Create from parent + modelServiceId; required and immutable on Update/Get/Delete
externalModelConfig GetAiGatewayModelServiceConfigRoutingFallbackDestinationExternalModelConfig: (ModelServiceConfigExternalModelConfig)
payPerTokenConfig GetAiGatewayModelServiceConfigRoutingFallbackDestinationPayPerTokenConfig: (ModelServiceConfigPayPerTokenConfig)
provisionedThroughputConfig GetAiGatewayModelServiceConfigRoutingFallbackDestinationProvisionedThroughputConfig: (ModelServiceConfigProvisionedThroughputConfig)
trafficPercentage Integer: (integer) - Share of traffic sent to this destination, 0-100. Optional on fallback destinations; see FallbackConfig

destinationType string: (string) - Backing-model category. Determines which oneof variant is populated. Possible values are: DESTINATION_TYPE_EXTERNAL_FOUNDATION_MODEL, DESTINATION_TYPE_PAY_PER_TOKEN_FOUNDATION_MODEL, DESTINATION_TYPE_PROVISIONED_THROUGHPUT_FOUNDATION_MODEL
isDeleted boolean: (boolean) - True when the destination's backing UC entity (MODEL for foundation-model destinations, MODEL_PROVIDER_SERVICE for external destinations) has been deleted but the destination row still references it. The dangling destination is surfaced (not silently dropped) so callers can see the broken routing. Inference traffic through this destination fails closed (BAD_REQUEST / FAILED_PRECONDITION)
name string: Resource name of the model service. Format: model-services/{catalog}.{schema}.{model_service}. Each {...} component is capped at 255 characters individually. Server-derived on Create from parent + modelServiceId; required and immutable on Update/Get/Delete
externalModelConfig GetAiGatewayModelServiceConfigRoutingFallbackDestinationExternalModelConfig: (ModelServiceConfigExternalModelConfig)
payPerTokenConfig GetAiGatewayModelServiceConfigRoutingFallbackDestinationPayPerTokenConfig: (ModelServiceConfigPayPerTokenConfig)
provisionedThroughputConfig GetAiGatewayModelServiceConfigRoutingFallbackDestinationProvisionedThroughputConfig: (ModelServiceConfigProvisionedThroughputConfig)
trafficPercentage number: (integer) - Share of traffic sent to this destination, 0-100. Optional on fallback destinations; see FallbackConfig

destination_type str: (string) - Backing-model category. Determines which oneof variant is populated. Possible values are: DESTINATION_TYPE_EXTERNAL_FOUNDATION_MODEL, DESTINATION_TYPE_PAY_PER_TOKEN_FOUNDATION_MODEL, DESTINATION_TYPE_PROVISIONED_THROUGHPUT_FOUNDATION_MODEL
is_deleted bool: (boolean) - True when the destination's backing UC entity (MODEL for foundation-model destinations, MODEL_PROVIDER_SERVICE for external destinations) has been deleted but the destination row still references it. The dangling destination is surfaced (not silently dropped) so callers can see the broken routing. Inference traffic through this destination fails closed (BAD_REQUEST / FAILED_PRECONDITION)
name str: Resource name of the model service. Format: model-services/{catalog}.{schema}.{model_service}. Each {...} component is capped at 255 characters individually. Server-derived on Create from parent + modelServiceId; required and immutable on Update/Get/Delete
external_model_config GetAiGatewayModelServiceConfigRoutingFallbackDestinationExternalModelConfig: (ModelServiceConfigExternalModelConfig)
pay_per_token_config GetAiGatewayModelServiceConfigRoutingFallbackDestinationPayPerTokenConfig: (ModelServiceConfigPayPerTokenConfig)
provisioned_throughput_config GetAiGatewayModelServiceConfigRoutingFallbackDestinationProvisionedThroughputConfig: (ModelServiceConfigProvisionedThroughputConfig)
traffic_percentage int: (integer) - Share of traffic sent to this destination, 0-100. Optional on fallback destinations; see FallbackConfig

destinationType String: (string) - Backing-model category. Determines which oneof variant is populated. Possible values are: DESTINATION_TYPE_EXTERNAL_FOUNDATION_MODEL, DESTINATION_TYPE_PAY_PER_TOKEN_FOUNDATION_MODEL, DESTINATION_TYPE_PROVISIONED_THROUGHPUT_FOUNDATION_MODEL
isDeleted Boolean: (boolean) - True when the destination's backing UC entity (MODEL for foundation-model destinations, MODEL_PROVIDER_SERVICE for external destinations) has been deleted but the destination row still references it. The dangling destination is surfaced (not silently dropped) so callers can see the broken routing. Inference traffic through this destination fails closed (BAD_REQUEST / FAILED_PRECONDITION)
name String: Resource name of the model service. Format: model-services/{catalog}.{schema}.{model_service}. Each {...} component is capped at 255 characters individually. Server-derived on Create from parent + modelServiceId; required and immutable on Update/Get/Delete
externalModelConfig Property Map: (ModelServiceConfigExternalModelConfig)
payPerTokenConfig Property Map: (ModelServiceConfigPayPerTokenConfig)
provisionedThroughputConfig Property Map: (ModelServiceConfigProvisionedThroughputConfig)
trafficPercentage Number: (integer) - Share of traffic sent to this destination, 0-100. Optional on fallback destinations; see FallbackConfig

GetAiGatewayModelServiceConfigRoutingFallbackDestinationExternalModelConfig

ModelProviderService string: (string) - Resource name of the governed ModelProviderService that owns provider auth and provider-specific configuration. The referenced ModelProviderService also carries the provider type, so this message does not surface it directly. Format: model-provider-services/{catalog}.{schema}.{model_provider_service}. Each {...} component is capped at 255 characters individually
Target GetAiGatewayModelServiceConfigRoutingFallbackDestinationExternalModelConfigTarget: (ModelProviderServiceConfigModelTargetConfig) - Routing target for the destination: the provider-side model selected from the referenced ModelProviderService's targets catalog, plus the unified API types the platform should translate to/from at request time

ModelProviderService string: (string) - Resource name of the governed ModelProviderService that owns provider auth and provider-specific configuration. The referenced ModelProviderService also carries the provider type, so this message does not surface it directly. Format: model-provider-services/{catalog}.{schema}.{model_provider_service}. Each {...} component is capped at 255 characters individually
Target GetAiGatewayModelServiceConfigRoutingFallbackDestinationExternalModelConfigTarget: (ModelProviderServiceConfigModelTargetConfig) - Routing target for the destination: the provider-side model selected from the referenced ModelProviderService's targets catalog, plus the unified API types the platform should translate to/from at request time

model_provider_service string: (string) - Resource name of the governed ModelProviderService that owns provider auth and provider-specific configuration. The referenced ModelProviderService also carries the provider type, so this message does not surface it directly. Format: model-provider-services/{catalog}.{schema}.{model_provider_service}. Each {...} component is capped at 255 characters individually
target object: (ModelProviderServiceConfigModelTargetConfig) - Routing target for the destination: the provider-side model selected from the referenced ModelProviderService's targets catalog, plus the unified API types the platform should translate to/from at request time

modelProviderService String: (string) - Resource name of the governed ModelProviderService that owns provider auth and provider-specific configuration. The referenced ModelProviderService also carries the provider type, so this message does not surface it directly. Format: model-provider-services/{catalog}.{schema}.{model_provider_service}. Each {...} component is capped at 255 characters individually
target GetAiGatewayModelServiceConfigRoutingFallbackDestinationExternalModelConfigTarget: (ModelProviderServiceConfigModelTargetConfig) - Routing target for the destination: the provider-side model selected from the referenced ModelProviderService's targets catalog, plus the unified API types the platform should translate to/from at request time

modelProviderService string: (string) - Resource name of the governed ModelProviderService that owns provider auth and provider-specific configuration. The referenced ModelProviderService also carries the provider type, so this message does not surface it directly. Format: model-provider-services/{catalog}.{schema}.{model_provider_service}. Each {...} component is capped at 255 characters individually
target GetAiGatewayModelServiceConfigRoutingFallbackDestinationExternalModelConfigTarget: (ModelProviderServiceConfigModelTargetConfig) - Routing target for the destination: the provider-side model selected from the referenced ModelProviderService's targets catalog, plus the unified API types the platform should translate to/from at request time

model_provider_service str: (string) - Resource name of the governed ModelProviderService that owns provider auth and provider-specific configuration. The referenced ModelProviderService also carries the provider type, so this message does not surface it directly. Format: model-provider-services/{catalog}.{schema}.{model_provider_service}. Each {...} component is capped at 255 characters individually
target GetAiGatewayModelServiceConfigRoutingFallbackDestinationExternalModelConfigTarget: (ModelProviderServiceConfigModelTargetConfig) - Routing target for the destination: the provider-side model selected from the referenced ModelProviderService's targets catalog, plus the unified API types the platform should translate to/from at request time

modelProviderService String: (string) - Resource name of the governed ModelProviderService that owns provider auth and provider-specific configuration. The referenced ModelProviderService also carries the provider type, so this message does not surface it directly. Format: model-provider-services/{catalog}.{schema}.{model_provider_service}. Each {...} component is capped at 255 characters individually
target Property Map: (ModelProviderServiceConfigModelTargetConfig) - Routing target for the destination: the provider-side model selected from the referenced ModelProviderService's targets catalog, plus the unified API types the platform should translate to/from at request time

GetAiGatewayModelServiceConfigRoutingFallbackDestinationExternalModelConfigTarget

Model string: (string) - UC model FQN of the model served by the backing endpoint (e.g., system.ai.databricks-claude-opus-4-6). Resolved from Model Serving at Create/Update time
NativeApiTypes List<string>: (list of string) - Provider-native API types the model supports (e.g. "openai/v1/chat/completions"). Used by the platform for request/response translation from the unified API type. At most 64 entries of at most 256 characters each; the list is persisted into the destination binding's bounded storage envelope

Model string: (string) - UC model FQN of the model served by the backing endpoint (e.g., system.ai.databricks-claude-opus-4-6). Resolved from Model Serving at Create/Update time
NativeApiTypes []string: (list of string) - Provider-native API types the model supports (e.g. "openai/v1/chat/completions"). Used by the platform for request/response translation from the unified API type. At most 64 entries of at most 256 characters each; the list is persisted into the destination binding's bounded storage envelope

model string: (string) - UC model FQN of the model served by the backing endpoint (e.g., system.ai.databricks-claude-opus-4-6). Resolved from Model Serving at Create/Update time
native_api_types list(string): (list of string) - Provider-native API types the model supports (e.g. "openai/v1/chat/completions"). Used by the platform for request/response translation from the unified API type. At most 64 entries of at most 256 characters each; the list is persisted into the destination binding's bounded storage envelope

model String: (string) - UC model FQN of the model served by the backing endpoint (e.g., system.ai.databricks-claude-opus-4-6). Resolved from Model Serving at Create/Update time
nativeApiTypes List<String>: (list of string) - Provider-native API types the model supports (e.g. "openai/v1/chat/completions"). Used by the platform for request/response translation from the unified API type. At most 64 entries of at most 256 characters each; the list is persisted into the destination binding's bounded storage envelope

model string: (string) - UC model FQN of the model served by the backing endpoint (e.g., system.ai.databricks-claude-opus-4-6). Resolved from Model Serving at Create/Update time
nativeApiTypes string[]: (list of string) - Provider-native API types the model supports (e.g. "openai/v1/chat/completions"). Used by the platform for request/response translation from the unified API type. At most 64 entries of at most 256 characters each; the list is persisted into the destination binding's bounded storage envelope

model str: (string) - UC model FQN of the model served by the backing endpoint (e.g., system.ai.databricks-claude-opus-4-6). Resolved from Model Serving at Create/Update time
native_api_types Sequence[str]: (list of string) - Provider-native API types the model supports (e.g. "openai/v1/chat/completions"). Used by the platform for request/response translation from the unified API type. At most 64 entries of at most 256 characters each; the list is persisted into the destination binding's bounded storage envelope

model String: (string) - UC model FQN of the model served by the backing endpoint (e.g., system.ai.databricks-claude-opus-4-6). Resolved from Model Serving at Create/Update time
nativeApiTypes List<String>: (list of string) - Provider-native API types the model supports (e.g. "openai/v1/chat/completions"). Used by the platform for request/response translation from the unified API type. At most 64 entries of at most 256 characters each; the list is persisted into the destination binding's bounded storage envelope

GetAiGatewayModelServiceConfigRoutingFallbackDestinationPayPerTokenConfig

Model string: (string) - UC model FQN of the model served by the backing endpoint (e.g., system.ai.databricks-claude-opus-4-6). Resolved from Model Serving at Create/Update time

Model string: (string) - UC model FQN of the model served by the backing endpoint (e.g., system.ai.databricks-claude-opus-4-6). Resolved from Model Serving at Create/Update time

model string: (string) - UC model FQN of the model served by the backing endpoint (e.g., system.ai.databricks-claude-opus-4-6). Resolved from Model Serving at Create/Update time

model String: (string) - UC model FQN of the model served by the backing endpoint (e.g., system.ai.databricks-claude-opus-4-6). Resolved from Model Serving at Create/Update time

model string: (string) - UC model FQN of the model served by the backing endpoint (e.g., system.ai.databricks-claude-opus-4-6). Resolved from Model Serving at Create/Update time

model str: (string) - UC model FQN of the model served by the backing endpoint (e.g., system.ai.databricks-claude-opus-4-6). Resolved from Model Serving at Create/Update time

model String: (string) - UC model FQN of the model served by the backing endpoint (e.g., system.ai.databricks-claude-opus-4-6). Resolved from Model Serving at Create/Update time

GetAiGatewayModelServiceConfigRoutingFallbackDestinationProvisionedThroughputConfig

Model string: (string) - UC model FQN of the model served by the backing endpoint (e.g., system.ai.databricks-claude-opus-4-6). Resolved from Model Serving at Create/Update time
ModelServingEndpoint string: (string) - Name of the backing Model Serving endpoint serving the provisioned- throughput foundation model, as the AIP-122 typed resource name serving-endpoints/{name}. The same UC model can be served on multiple Model Serving endpoints (different throughput / region / config); the caller picks which one this destination routes to. The endpoint must exist at create time

Model string: (string) - UC model FQN of the model served by the backing endpoint (e.g., system.ai.databricks-claude-opus-4-6). Resolved from Model Serving at Create/Update time
ModelServingEndpoint string: (string) - Name of the backing Model Serving endpoint serving the provisioned- throughput foundation model, as the AIP-122 typed resource name serving-endpoints/{name}. The same UC model can be served on multiple Model Serving endpoints (different throughput / region / config); the caller picks which one this destination routes to. The endpoint must exist at create time

model string: (string) - UC model FQN of the model served by the backing endpoint (e.g., system.ai.databricks-claude-opus-4-6). Resolved from Model Serving at Create/Update time
model_serving_endpoint string: (string) - Name of the backing Model Serving endpoint serving the provisioned- throughput foundation model, as the AIP-122 typed resource name serving-endpoints/{name}. The same UC model can be served on multiple Model Serving endpoints (different throughput / region / config); the caller picks which one this destination routes to. The endpoint must exist at create time

model String: (string) - UC model FQN of the model served by the backing endpoint (e.g., system.ai.databricks-claude-opus-4-6). Resolved from Model Serving at Create/Update time
modelServingEndpoint String: (string) - Name of the backing Model Serving endpoint serving the provisioned- throughput foundation model, as the AIP-122 typed resource name serving-endpoints/{name}. The same UC model can be served on multiple Model Serving endpoints (different throughput / region / config); the caller picks which one this destination routes to. The endpoint must exist at create time

model string: (string) - UC model FQN of the model served by the backing endpoint (e.g., system.ai.databricks-claude-opus-4-6). Resolved from Model Serving at Create/Update time
modelServingEndpoint string: (string) - Name of the backing Model Serving endpoint serving the provisioned- throughput foundation model, as the AIP-122 typed resource name serving-endpoints/{name}. The same UC model can be served on multiple Model Serving endpoints (different throughput / region / config); the caller picks which one this destination routes to. The endpoint must exist at create time

model str: (string) - UC model FQN of the model served by the backing endpoint (e.g., system.ai.databricks-claude-opus-4-6). Resolved from Model Serving at Create/Update time
model_serving_endpoint str: (string) - Name of the backing Model Serving endpoint serving the provisioned- throughput foundation model, as the AIP-122 typed resource name serving-endpoints/{name}. The same UC model can be served on multiple Model Serving endpoints (different throughput / region / config); the caller picks which one this destination routes to. The endpoint must exist at create time

model String: (string) - UC model FQN of the model served by the backing endpoint (e.g., system.ai.databricks-claude-opus-4-6). Resolved from Model Serving at Create/Update time
modelServingEndpoint String: (string) - Name of the backing Model Serving endpoint serving the provisioned- throughput foundation model, as the AIP-122 typed resource name serving-endpoints/{name}. The same UC model can be served on multiple Model Serving endpoints (different throughput / region / config); the caller picks which one this destination routes to. The endpoint must exist at create time

GetAiGatewayModelServiceProviderConfig

WorkspaceId string: Workspace ID which the resource belongs to. This workspace must be part of the account which the provider is configured with.

WorkspaceId string: Workspace ID which the resource belongs to. This workspace must be part of the account which the provider is configured with.

workspace_id string: Workspace ID which the resource belongs to. This workspace must be part of the account which the provider is configured with.

workspaceId String: Workspace ID which the resource belongs to. This workspace must be part of the account which the provider is configured with.

workspaceId string: Workspace ID which the resource belongs to. This workspace must be part of the account which the provider is configured with.

workspace_id str: Workspace ID which the resource belongs to. This workspace must be part of the account which the provider is configured with.

workspaceId String: Workspace ID which the resource belongs to. This workspace must be part of the account which the provider is configured with.

Package Details

Repository: databricks pulumi/pulumi-databricks
License: Apache-2.0
Notes: This Pulumi package is based on the databricks Terraform Provider.

Viewing docs for Databricks v1.103.0
published on Tuesday, Aug 11, 2026 by Pulumi

Schema (JSON)

pulumi/pulumi-databricks

databricks.getAiGatewayModelService

On this page

On this page

Using getAiGatewayModelService

getAiGatewayModelService Result

Supporting Types

GetAiGatewayModelServiceConfig

GetAiGatewayModelServiceConfigInferenceTable

GetAiGatewayModelServiceConfigRateLimit

GetAiGatewayModelServiceConfigRouting

GetAiGatewayModelServiceConfigRoutingDestination

GetAiGatewayModelServiceConfigRoutingDestinationExternalModelConfig

GetAiGatewayModelServiceConfigRoutingDestinationExternalModelConfigTarget

GetAiGatewayModelServiceConfigRoutingDestinationPayPerTokenConfig

GetAiGatewayModelServiceConfigRoutingDestinationProvisionedThroughputConfig

GetAiGatewayModelServiceConfigRoutingFallback

GetAiGatewayModelServiceConfigRoutingFallbackDestination

GetAiGatewayModelServiceConfigRoutingFallbackDestinationExternalModelConfig

GetAiGatewayModelServiceConfigRoutingFallbackDestinationExternalModelConfigTarget

GetAiGatewayModelServiceConfigRoutingFallbackDestinationPayPerTokenConfig

GetAiGatewayModelServiceConfigRoutingFallbackDestinationProvisionedThroughputConfig

GetAiGatewayModelServiceProviderConfig

Package Details

On this page

On this page

Try Pulumi Cloud free.
Your team will thank you.

databricks.getAiGatewayModelService

On this page

On this page

Using getAiGatewayModelService

getAiGatewayModelService Result

Supporting Types

GetAiGatewayModelServiceConfig

GetAiGatewayModelServiceConfigInferenceTable

GetAiGatewayModelServiceConfigRateLimit

GetAiGatewayModelServiceConfigRouting

GetAiGatewayModelServiceConfigRoutingDestination

GetAiGatewayModelServiceConfigRoutingDestinationExternalModelConfig

GetAiGatewayModelServiceConfigRoutingDestinationExternalModelConfigTarget

GetAiGatewayModelServiceConfigRoutingDestinationPayPerTokenConfig

GetAiGatewayModelServiceConfigRoutingDestinationProvisionedThroughputConfig

GetAiGatewayModelServiceConfigRoutingFallback

GetAiGatewayModelServiceConfigRoutingFallbackDestination

GetAiGatewayModelServiceConfigRoutingFallbackDestinationExternalModelConfig

GetAiGatewayModelServiceConfigRoutingFallbackDestinationExternalModelConfigTarget

GetAiGatewayModelServiceConfigRoutingFallbackDestinationPayPerTokenConfig

GetAiGatewayModelServiceConfigRoutingFallbackDestinationProvisionedThroughputConfig

GetAiGatewayModelServiceProviderConfig

Package Details

On this page

On this page

Try Pulumi Cloud free.Your team will thank you.

Try Pulumi Cloud free.
Your team will thank you.