AiGatewayModelService

laktory.models.resources.databricks.AiGatewayModelService

Bases: AiGatewayModelServiceBase

Databricks AI Gateway model service

Stable, Unity Catalog governed endpoint routing inference requests to one or more destinations (Databricks foundation models, provisioned throughput or external model provider services), with rate limits, fallbacks and inference tables.

Examples:

import io

from laktory import models

service_yaml = '''
parent: schemas/dev.ai
model_service_id: customer_support
comment: Routes customer support requests
config:
  routing:
    destinations:
    - name: primary
      destination_type: DESTINATION_TYPE_PAY_PER_TOKEN_FOUNDATION_MODEL
      pay_per_token_config:
        model: models/system.ai.databricks-gpt-5
      traffic_percentage: 100
'''
service = models.resources.databricks.AiGatewayModelService.model_validate_yaml(
    io.StringIO(service_yaml)
)

References

• Databricks AI Gateway
• Terraform databricks_ai_gateway_model_service

BASE	DESCRIPTION
comment	User-provided description TYPE: str | None | VariableType DEFAULT: None
config	Destinations, routing, rate limits, and payload logging configuration. Required on Create. On Update, provide this field when update_mask contains config or one of its subpaths TYPE: AiGatewayModelServiceConfig | None | VariableType DEFAULT: None
model_service_id	Name for the model service, e.g. 'my_model_service' TYPE: str | VariableType
parent	Parent Unity Catalog schema where the inference table is created, in the form schemas/{catalog}.{schema}. Required when configuring an inference table. After the inference table is created, this field cannot be changed TYPE: str | VariableType
provider_config	Configure the provider for management through account provider TYPE: AiGatewayModelServiceProviderConfig | None | VariableType DEFAULT: None

---

laktory.models.resources.databricks.aigatewaymodelservice.AiGatewayModelServiceConfig

Bases: BaseModel

PARAMETER	DESCRIPTION
inference_table	Inference table configuration for payload logging TYPE: AiGatewayModelServiceConfigInferenceTable | None | VariableType DEFAULT: None
rate_limits	Rate limits applied to requests routed through this model service TYPE: list[AiGatewayModelServiceConfigRateLimits] | None | VariableType DEFAULT: None
routing	Routing configuration: destinations and fallback TYPE: AiGatewayModelServiceConfigRouting | None | VariableType DEFAULT: None

---

laktory.models.resources.databricks.aigatewaymodelservice.AiGatewayModelServiceConfigInferenceTable

Bases: BaseModel

PARAMETER	DESCRIPTION
parent	Parent Unity Catalog schema where the inference table is created, in the form schemas/{catalog}.{schema}. Required when configuring an inference table. After the inference table is created, this field cannot be changed TYPE: str | VariableType
table_name_prefix	Prefix used to form the inference table's registered name. AI Gateway appends _payload; for example, table_name_prefix = 'orders' creates orders_payload. If unset, the prefix defaults to the service name. Read table from the response for the resulting resource name. After the inference table is created, this field cannot be changed TYPE: str | None | VariableType DEFAULT: None

---

laktory.models.resources.databricks.aigatewaymodelservice.AiGatewayModelServiceConfigRateLimits

Bases: BaseModel

PARAMETER	DESCRIPTION
key	Scope of the rate limit. Depending on this value, the limit applies to a principal, the service as a whole, or each user by default. Possible values are: RATE_LIMIT_KEY_SERVICE, RATE_LIMIT_KEY_SERVICE_PRINCIPAL, RATE_LIMIT_KEY_USER, RATE_LIMIT_KEY_USER_DEFAULT, RATE_LIMIT_KEY_USER_GROUP TYPE: str | VariableType
principal	Principal this limit applies to: user email, group name, or service principal application ID. Required when key applies to a user, group, or service principal; otherwise it must be unset TYPE: str | None | VariableType DEFAULT: None
renewal_period	Renewal period. Possible values are: RATE_LIMIT_RENEWAL_PERIOD_HOUR, RATE_LIMIT_RENEWAL_PERIOD_MINUTE TYPE: str | VariableType
requests	Maximum requests allowed in one renewal period. Leave unset for no request limit. Set to 0 to deny all requests TYPE: int | None | VariableType DEFAULT: None
tokens	Maximum tokens allowed in one renewal period. Leave unset for no token limit. Set to 0 to deny all requests TYPE: int | None | VariableType DEFAULT: None

---

laktory.models.resources.databricks.aigatewaymodelservice.AiGatewayModelServiceConfigRouting

Bases: BaseModel

PARAMETER	DESCRIPTION
destinations	Primary routing destinations. At most 10 are allowed. At least one is required on Create. On Update, provide this list when replacing the full config or updating config.routing.destinations; other granular routing updates do not require resending destinations. The intermediate config.routing mask path is not supported TYPE: list[AiGatewayModelServiceConfigRoutingDestinations] | None | VariableType DEFAULT: None
fallback	Fallback routing applied after a primary destination fails. Fallback destinations are tried in the listed order TYPE: AiGatewayModelServiceConfigRoutingFallback | None | VariableType DEFAULT: None

---

laktory.models.resources.databricks.aigatewaymodelservice.AiGatewayModelServiceConfigRoutingDestinations

Bases: BaseModel

PARAMETER	DESCRIPTION
destination_type	Backing-model category. Provide the matching type-specific configuration and leave the other type-specific configurations unset. Possible values are: DESTINATION_TYPE_EXTERNAL_FOUNDATION_MODEL, DESTINATION_TYPE_PAY_PER_TOKEN_FOUNDATION_MODEL, DESTINATION_TYPE_PROVISIONED_THROUGHPUT_FOUNDATION_MODEL TYPE: str | VariableType
external_model_config	Configuration for an external model reached through a model provider service TYPE: AiGatewayModelServiceConfigRoutingDestinationsExternalModelConfig | None | VariableType DEFAULT: None
name	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 + model_service_id; required and immutable on Update/Get/Delete TYPE: str | VariableType
pay_per_token_config	Configuration for a pay-per-token Databricks foundation model TYPE: AiGatewayModelServiceConfigRoutingDestinationsPayPerTokenConfig | None | VariableType DEFAULT: None
provisioned_throughput_config	Configuration for a provisioned-throughput Databricks foundation model TYPE: AiGatewayModelServiceConfigRoutingDestinationsProvisionedThroughputConfig | None | VariableType DEFAULT: None
traffic_percentage	Percentage of primary traffic sent to this destination, from 0 to 100. Required when there is more than one primary destination, in which case the primary percentages must sum to 100; a single primary destination receives all traffic. Fallback destinations are ordered and do not use this field TYPE: int | None | VariableType DEFAULT: None

---

laktory.models.resources.databricks.aigatewaymodelservice.AiGatewayModelServiceConfigRoutingDestinationsExternalModelConfig

Bases: BaseModel

PARAMETER	DESCRIPTION
model_provider_service	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 TYPE: str | VariableType
target	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 TYPE: AiGatewayModelServiceConfigRoutingDestinationsExternalModelConfigTarget | None | VariableType DEFAULT: None

---

laktory.models.resources.databricks.aigatewaymodelservice.AiGatewayModelServiceConfigRoutingDestinationsExternalModelConfigTarget

Bases: BaseModel

PARAMETER	DESCRIPTION
model	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 TYPE: str | VariableType
native_api_types	Provider-native API types supported by this model, such as openai/v1/chat/completions. At least one value is required. AI Gateway uses these values to translate requests and responses. At most 64 entries of 256 characters each are allowed TYPE: list[str] | None | VariableType DEFAULT: None

---

laktory.models.resources.databricks.aigatewaymodelservice.AiGatewayModelServiceConfigRoutingDestinationsPayPerTokenConfig

Bases: BaseModel

PARAMETER	DESCRIPTION
model	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 TYPE: str | VariableType

---

laktory.models.resources.databricks.aigatewaymodelservice.AiGatewayModelServiceConfigRoutingDestinationsProvisionedThroughputConfig

Bases: BaseModel

PARAMETER	DESCRIPTION
model_serving_endpoint	Name of the backing Model Serving endpoint serving the provisioned- throughput foundation model, in the form serving-endpoints/{name}. The same Unity Catalog model can be served on multiple Model Serving endpoints with different throughput, regions, or configurations. The caller selects the endpoint to which this destination routes. The endpoint must exist at create time TYPE: str | VariableType

---

laktory.models.resources.databricks.aigatewaymodelservice.AiGatewayModelServiceConfigRoutingFallback

Bases: BaseModel

PARAMETER

DESCRIPTION

destinations

Primary routing destinations. At most 10 are allowed. At least one is required on Create. On Update, provide this list when replacing the full config or updating config.routing.destinations; other granular routing updates do not require resending destinations. The intermediate config.routing mask path is not supported TYPE: list[AiGatewayModelServiceConfigRoutingFallbackDestinations] | None | VariableType DEFAULT: None

---

laktory.models.resources.databricks.aigatewaymodelservice.AiGatewayModelServiceConfigRoutingFallbackDestinations

Bases: BaseModel

PARAMETER	DESCRIPTION
destination_type	Backing-model category. Provide the matching type-specific configuration and leave the other type-specific configurations unset. Possible values are: DESTINATION_TYPE_EXTERNAL_FOUNDATION_MODEL, DESTINATION_TYPE_PAY_PER_TOKEN_FOUNDATION_MODEL, DESTINATION_TYPE_PROVISIONED_THROUGHPUT_FOUNDATION_MODEL TYPE: str | VariableType
external_model_config	Configuration for an external model reached through a model provider service TYPE: AiGatewayModelServiceConfigRoutingFallbackDestinationsExternalModelConfig | None | VariableType DEFAULT: None
name	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 + model_service_id; required and immutable on Update/Get/Delete TYPE: str | VariableType
pay_per_token_config	Configuration for a pay-per-token Databricks foundation model TYPE: AiGatewayModelServiceConfigRoutingFallbackDestinationsPayPerTokenConfig | None | VariableType DEFAULT: None
provisioned_throughput_config	Configuration for a provisioned-throughput Databricks foundation model TYPE: AiGatewayModelServiceConfigRoutingFallbackDestinationsProvisionedThroughputConfig | None | VariableType DEFAULT: None
traffic_percentage	Percentage of primary traffic sent to this destination, from 0 to 100. Required when there is more than one primary destination, in which case the primary percentages must sum to 100; a single primary destination receives all traffic. Fallback destinations are ordered and do not use this field TYPE: int | None | VariableType DEFAULT: None

---

laktory.models.resources.databricks.aigatewaymodelservice.AiGatewayModelServiceConfigRoutingFallbackDestinationsExternalModelConfig

Bases: BaseModel

PARAMETER	DESCRIPTION
model_provider_service	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 TYPE: str | VariableType
target	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 TYPE: AiGatewayModelServiceConfigRoutingFallbackDestinationsExternalModelConfigTarget | None | VariableType DEFAULT: None

---

laktory.models.resources.databricks.aigatewaymodelservice.AiGatewayModelServiceConfigRoutingFallbackDestinationsExternalModelConfigTarget

Bases: BaseModel

PARAMETER	DESCRIPTION
model	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 TYPE: str | VariableType
native_api_types	Provider-native API types supported by this model, such as openai/v1/chat/completions. At least one value is required. AI Gateway uses these values to translate requests and responses. At most 64 entries of 256 characters each are allowed TYPE: list[str] | None | VariableType DEFAULT: None

---

laktory.models.resources.databricks.aigatewaymodelservice.AiGatewayModelServiceConfigRoutingFallbackDestinationsPayPerTokenConfig

Bases: BaseModel

PARAMETER	DESCRIPTION
model	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 TYPE: str | VariableType

---

laktory.models.resources.databricks.aigatewaymodelservice.AiGatewayModelServiceConfigRoutingFallbackDestinationsProvisionedThroughputConfig

Bases: BaseModel

PARAMETER	DESCRIPTION
model_serving_endpoint	Name of the backing Model Serving endpoint serving the provisioned- throughput foundation model, in the form serving-endpoints/{name}. The same Unity Catalog model can be served on multiple Model Serving endpoints with different throughput, regions, or configurations. The caller selects the endpoint to which this destination routes. The endpoint must exist at create time TYPE: str | VariableType

---

laktory.models.resources.databricks.aigatewaymodelservice.AiGatewayModelServiceProviderConfig

Bases: BaseModel

PARAMETER	DESCRIPTION
workspace_id	Workspace ID which the resource belongs to. This workspace must be part of the account which the provider is configured with TYPE: str | None | VariableType DEFAULT: None