Model Provider Services

Package: databricks.bundles.model_provider_services

Classes

class InferenceTableConfig

Configuration for logging request and response payloads to a Unity Catalog inference table. When this configuration is present, payload logging is enabled by default.

parent: str

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.

table_name_prefix: str | None = None

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.

classmethod from_dict(

value: dict,

) → Self

as_dict(

self,

) → dict

class Lifecycle

prevent_destroy: bool | None = None

Lifecycle setting to prevent the resource from being destroyed.

classmethod from_dict(

value: dict,

) → Self

as_dict(

self,

) → dict

class ModelProviderService

model_provider_service_id: str

parent: str

comment: str | None = None

config: ModelProviderServiceConfig | None = None

grants: list[PrivilegeAssignment]

lifecycle: Lifecycle | None = None

classmethod from_dict(

value: dict,

) → Self

as_dict(

self,

) → dict

class ModelProviderServiceConfig

Behavioral configuration for a ModelProviderService: provider authentication and provider-specific fields, the catalog of models this provider service can route to, and the passthrough policy that governs how request headers, query parameters, and unmanaged subpaths cross the trust boundary to the upstream provider.

allow_all_targets: bool | None = None

When true, accepts any model exposed by the upstream provider; targets is not required and does not restrict routability. When false, only models listed in targets are routable. Defaults to false.

amazon_bedrock: ModelProviderServiceConfigAmazonBedrockProviderConfig | None = None

Amazon Bedrock provider configuration.

anthropic: ModelProviderServiceConfigAnthropicProviderConfig | None = None

Anthropic provider configuration. Exactly one of direct or relayed must be set on Create; the two are mutually exclusive.

azure_openai: ModelProviderServiceConfigAzureOpenAiProviderConfig | None = None

Azure OpenAI provider configuration.

custom: ModelProviderServiceConfigCustomProviderConfig | None = None

Custom OpenAI-compatible provider configuration with bearer-token authentication.

forward_headers: bool | None = None

Whether to forward incoming HTTP headers to the upstream provider. Defaults to false and is configured for the entire provider service, not per request. Upstream authentication is configured separately in the provider-specific configuration.

forward_query_parameters: bool | None = None

Whether to forward incoming query parameters to the upstream provider. Defaults to false and is configured for the entire provider service, not per request.

forward_unmanaged_paths: bool | None = None

Whether to proxy paths that AI Gateway does not recognize as configured provider-native API types. Defaults to false. When true, these paths are forwarded unchanged to the upstream provider. When false, only recognized API paths are served. Enabling this broadens the upstream API surface exposed through the provider service.

gemini_enterprise: ModelProviderServiceConfigGeminiEnterpriseProviderConfig | None = None

Gemini Enterprise provider configuration.

inference_table: InferenceTableConfig | None = None

Payload logging configuration for requests sent directly to this provider service. Requests routed through a model service are captured by that model service’s inference table instead.

microsoft_foundry: ModelProviderServiceConfigMicrosoftFoundryProviderConfig | None = None

Microsoft Foundry provider configuration.

openai: ModelProviderServiceConfigOpenAiProviderConfig | None = None

OpenAI provider configuration.

provider_type: ModelProviderServiceConfigExternalModelProviderType | None = None

External model provider. Required on Create and immutable thereafter. Set the matching provider-specific configuration, such as openai, azure_openai, or amazon_bedrock.

rate_limits: list[RateLimit]

Rate limits for requests sent directly to this provider service. Requests routed through a model service use that model service’s rate limits instead.

targets: list[ModelProviderServiceConfigModelTargetConfig]

Models and provider-native API types exposed by this provider service. Each entry must include at least one native_api_types value. When allow_all_targets is false, at least one entry is required and model service destinations can reference only listed models. When allow_all_targets is true, any upstream model is routable; entries in this list provide API-type metadata without restricting other models.

classmethod from_dict(

value: dict,

) → Self

as_dict(

self,

) → dict

class ModelProviderServiceConfigAmazonBedrockProviderConfig

Amazon Bedrock provider configuration.

direct: ModelProviderServiceConfigAmazonBedrockProviderDirectConfig | None = None

Amazon Bedrock region and authentication configuration.

classmethod from_dict(

value: dict,

) → Self

as_dict(

self,

) → dict

class ModelProviderServiceConfigAmazonBedrockProviderDirectConfig

Direct form of Amazon Bedrock provider config.

Authentication is one of two mutually exclusive modes, exactly one of which must be supplied on Create: - Access keys: set aws_access_key, leave service_credential unset. - Unity Catalog service credential: set service_credential.name to the resource name credentials/{name}, leave aws_access_key unset. The credential value lives in Unity Catalog and is referenced by name, not held on this message. Setting more than one mode is rejected.

aws_access_key: ModelProviderServiceConfigAwsAccessKey | None = None

AWS access-key-pair authentication. Set access_key_id and secret_access_key.plaintext. Mutually exclusive with service_credential.

region: str | None = None

AWS region where the Bedrock endpoint is hosted (e.g., us-east-1). Required on Create.

service_credential: ModelProviderServiceConfigServiceCredential | None = None

Reference to a Unity Catalog service credential authorizing Bedrock requests. On Create, supply service_credential.name in the form credentials/{name}. Required on Create when using service-credential authentication; mutually exclusive with aws_access_key. The credential is referenced by name; its value is not carried here. Only supported on AWS-hosted workspaces.

classmethod from_dict(

value: dict,

) → Self

as_dict(

self,

) → dict

class ModelProviderServiceConfigAnthropicProviderConfig

Anthropic provider configuration. Exactly one of direct or relayed must be set on Create; the two are mutually exclusive.

direct: ModelProviderServiceConfigAnthropicProviderDirectConfig | None = None

Direct authentication with an API key supplied in direct.api_key.plaintext. Required unless relayed is set.

relayed: ModelProviderServiceConfigAnthropicProviderRelayedConfig | None = None

Relayed authentication. Each inference request supplies the caller’s OAuth token, which is forwarded to Anthropic. No Anthropic credential is stored. Mutually exclusive with direct.

classmethod from_dict(

value: dict,

) → Self

as_dict(

self,

) → dict

class ModelProviderServiceConfigAnthropicProviderDirectConfig

Direct form of Anthropic provider config.

api_key: ModelProviderServiceConfigProviderSecret | None = None

Anthropic API key. Required when creating the service. Supply the value in api_key.plaintext.

classmethod from_dict(

value: dict,

) → Self

as_dict(

self,

) → dict

class ModelProviderServiceConfigAnthropicProviderRelayedConfig

Relayed Anthropic provider configuration. Each inference request supplies the caller’s OAuth token, which is forwarded to Anthropic. No Anthropic credential is stored.

classmethod from_dict(

value: dict,

) → Self

as_dict(

self,

) → dict

class ModelProviderServiceConfigAwsAccessKey

AWS access-key-pair auth for Amazon Bedrock: a SigV4-signing key pair.

access_key_id: str | None = None

AWS access key ID. Required on Create when using access-key auth. Treated as username-equivalent (not a secret value): round-trips on reads and is scrubbed from audit logs.

secret_access_key: ModelProviderServiceConfigProviderSecret | None = None

AWS secret access key paired with access_key_id. Required when creating a service with access-key authentication. Supply the value in secret_access_key.plaintext.

classmethod from_dict(

value: dict,

) → Self

as_dict(

self,

) → dict

class ModelProviderServiceConfigAzureOpenAiProviderConfig

Azure OpenAI provider configuration.

direct: ModelProviderServiceConfigAzureOpenAiProviderDirectConfig | None = None

Azure OpenAI endpoint and authentication configuration.

classmethod from_dict(

value: dict,

) → Self

as_dict(

self,

) → dict

class ModelProviderServiceConfigAzureOpenAiProviderDirectConfig

Direct form of Azure OpenAI provider config. Exactly one of three mutually-exclusive auth modes must be supplied on Create: - API key: set api_key, leave entra_service_principal and service_credential unset. - Entra ID (service principal): set entra_service_principal, leave api_key and service_credential unset. - Unity Catalog service credential: set service_credential.name to the resource name credentials/{name}, leave api_key and entra_service_principal unset. The credential value lives in Unity Catalog and is referenced by name, not held on this message. Only supported on Azure-hosted workspaces. Setting more than one mode is rejected.

api_key: ModelProviderServiceConfigProviderSecret | None = None

Azure OpenAI API key. Supply the value in api_key.plaintext. Mutually exclusive with Entra ID and Unity Catalog service credential authentication.

base_url: str | None = None

Full Azure OpenAI endpoint base URL, e.g. https://myresource.openai.azure.com. Required on Create.

entra_service_principal: ModelProviderServiceConfigEntraServicePrincipal | None = None

Entra ID service-principal authentication. Set tenant_id, client_id, and client_secret.plaintext. Mutually exclusive with api_key and service_credential.

service_credential: ModelProviderServiceConfigServiceCredential | None = None

Reference to a Unity Catalog service credential authorizing Azure OpenAI requests. On Create, supply service_credential.name in the form credentials/{name}. Required on Create when using service-credential authentication; mutually exclusive with api_key and entra_service_principal. The credential is referenced by name; its value is not carried here. Only supported on Azure-hosted workspaces.

classmethod from_dict(

value: dict,

) → Self

as_dict(

self,

) → dict

class ModelProviderServiceConfigCustomProviderConfig

Custom OpenAI-compatible provider configuration with bearer-token authentication.

direct: ModelProviderServiceConfigCustomProviderDirectConfig | None = None

Endpoint and authentication configuration for the custom provider.

classmethod from_dict(

value: dict,

) → Self

as_dict(

self,

) → dict

class ModelProviderServiceConfigCustomProviderDirectConfig

Direct form of a custom provider configuration. Set api_key to send the secret as an Authorization bearer token, or header_auth to forward it under a caller-chosen HTTP header.

api_key: ModelProviderServiceConfigProviderSecret | None = None

Bearer token forwarded in the Authorization header. Supply the value in api_key.plaintext.

base_url: str | None = None

Endpoint URL of the OpenAI-compatible service (e.g., https://api.example.com/v1). Required on Create.

classmethod from_dict(

value: dict,

) → Self

as_dict(

self,

) → dict

class ModelProviderServiceConfigEntraServicePrincipal

Entra ID (Azure AD) service-principal authentication. The tenant_id and client_id identify the service principal, and client_secret authenticates it. AI Gateway exchanges these credentials for an Entra bearer token for requests to Azure OpenAI or Microsoft Foundry.

client_id: str | None = None

Entra ID client (application) ID. Required on Create.

client_secret: ModelProviderServiceConfigProviderSecret | None = None

Entra ID client secret. Supply the value in client_secret.plaintext.

tenant_id: str | None = None

Entra ID (Azure AD) tenant ID. Required on Create.

classmethod from_dict(

value: dict,

) → Self

as_dict(

self,

) → dict

class ModelProviderServiceConfigExternalModelProviderType

External LLM provider for an EXTERNAL_FOUNDATION_MODEL destination.

EXTERNAL_MODEL_PROVIDER_TYPE_OPENAI = 'EXTERNAL_MODEL_PROVIDER_TYPE_OPENAI'

EXTERNAL_MODEL_PROVIDER_TYPE_AZURE_OPENAI = 'EXTERNAL_MODEL_PROVIDER_TYPE_AZURE_OPENAI'

EXTERNAL_MODEL_PROVIDER_TYPE_ANTHROPIC = 'EXTERNAL_MODEL_PROVIDER_TYPE_ANTHROPIC'

EXTERNAL_MODEL_PROVIDER_TYPE_AMAZON_BEDROCK = 'EXTERNAL_MODEL_PROVIDER_TYPE_AMAZON_BEDROCK'

EXTERNAL_MODEL_PROVIDER_TYPE_CUSTOM = 'EXTERNAL_MODEL_PROVIDER_TYPE_CUSTOM'

EXTERNAL_MODEL_PROVIDER_TYPE_MICROSOFT_FOUNDRY = 'EXTERNAL_MODEL_PROVIDER_TYPE_MICROSOFT_FOUNDRY'

EXTERNAL_MODEL_PROVIDER_TYPE_GEMINI_ENTERPRISE = 'EXTERNAL_MODEL_PROVIDER_TYPE_GEMINI_ENTERPRISE'

class ModelProviderServiceConfigGeminiEnterpriseProviderConfig

Gemini Enterprise provider configuration.

direct: ModelProviderServiceConfigGeminiEnterpriseProviderDirectConfig | None = None

Gemini Enterprise project, region, and authentication configuration.

classmethod from_dict(

value: dict,

) → Self

as_dict(

self,

) → dict

class ModelProviderServiceConfigGeminiEnterpriseProviderDirectConfig

Direct Gemini Enterprise provider configuration. An API key is required when creating the service.

api_key: ModelProviderServiceConfigProviderSecret | None = None

Google Gemini Enterprise API key. Required when creating the service. Supply the value in api_key.plaintext.

project_id: str | None = None

GCP project ID hosting the Gemini Enterprise endpoint. Required on Create.

region: str | None = None

GCP region of the Gemini Enterprise endpoint (e.g., us-central1). Required on Create.

classmethod from_dict(

value: dict,

) → Self

as_dict(

self,

) → dict

class ModelProviderServiceConfigMicrosoftFoundryProviderConfig

Microsoft Foundry provider configuration.

direct: ModelProviderServiceConfigMicrosoftFoundryProviderDirectConfig | None = None

Microsoft Foundry endpoint and authentication configuration.

classmethod from_dict(

value: dict,

) → Self

as_dict(

self,

) → dict

class ModelProviderServiceConfigMicrosoftFoundryProviderDirectConfig

Direct form of Microsoft Foundry provider config.

Authentication is one of three mutually exclusive modes, exactly one of which must be supplied on Create: - API key: set api_key, leave entra_service_principal and service_credential unset. - Entra ID (service principal): set entra_service_principal, leave api_key and service_credential unset. AI Gateway exchanges these for an Entra bearer token on outbound requests via the OAuth2 client-credentials grant. - Unity Catalog service credential: set service_credential.name to the resource name credentials/{name}, leave api_key and entra_service_principal unset. The credential value lives in Unity Catalog and is referenced by name, not held on this message. Only supported on Azure-hosted workspaces. Setting more than one mode is rejected.

api_key: ModelProviderServiceConfigProviderSecret | None = None

Microsoft Foundry API key. Supply the value in api_key.plaintext. Mutually exclusive with Entra ID and Unity Catalog service credential authentication.

base_url: str | None = None

Microsoft Foundry endpoint URL. Required on Create.

entra_service_principal: ModelProviderServiceConfigEntraServicePrincipal | None = None

Entra ID service-principal authentication. Set tenant_id, client_id, and client_secret.plaintext. Mutually exclusive with api_key and service_credential.

service_credential: ModelProviderServiceConfigServiceCredential | None = None

Reference to a Unity Catalog service credential authorizing Microsoft Foundry requests. On Create, supply service_credential.name in the form credentials/{name}. Required on Create when using service-credential authentication; mutually exclusive with api_key and entra_service_principal. The credential is referenced by name; its value is not carried here. Only supported on Azure-hosted workspaces.

classmethod from_dict(

value: dict,

) → Self

as_dict(

self,

) → dict

class ModelProviderServiceConfigModelTargetConfig

Model target configuration for an external model destination.

model: str

Provider-side model identifier, such as gpt-5 or claude-opus-4-7. This identifies a model at the upstream provider; it is not a Unity Catalog model resource.

native_api_types: list[str]

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.

classmethod from_dict(

value: dict,

) → Self

as_dict(

self,

) → dict

class ModelProviderServiceConfigOpenAiProviderConfig

OpenAI provider configuration.

direct: ModelProviderServiceConfigOpenAiProviderDirectConfig | None = None

OpenAI configuration with an API key supplied in the request.

classmethod from_dict(

value: dict,

) → Self

as_dict(

self,

) → dict

class ModelProviderServiceConfigOpenAiProviderDirectConfig

Direct (inline-credentials) form of the OpenAI provider config.

api_key: ModelProviderServiceConfigProviderSecret | None = None

OpenAI API key. Required when creating the service. Supply the value in api_key.plaintext.

base_url: str | None = None

Optional custom base URL. Defaults to https://api.openai.com/v1. Use for OpenAI-API-compatible third-party endpoints or in-network proxies.

organization: str | None = None

Optional OpenAI organization ID. When set, the platform forwards it as the OpenAI-Organization header.

classmethod from_dict(

value: dict,

) → Self

as_dict(

self,

) → dict

class ModelProviderServiceConfigProviderSecret

A secret value supplied as part of an inline provider config. The caller supplies the value as inline plaintext on writes; the platform stores it encrypted. The plaintext field is INPUT_ONLY and never round-trips on reads.

plaintext: str | None = None

Inline plaintext credential. INPUT_ONLY: the value never round-trips on reads. Get and List responses omit plaintext; the enclosing secret object remains present to indicate that a secret is configured.

classmethod from_dict(

value: dict,

) → Self

as_dict(

self,

) → dict

class ModelProviderServiceConfigServiceCredential

The customer-owned Unity Catalog service credential a model provider service uses to authenticate to its provider, referenced by name.

name: str

Resource name of the bound Unity Catalog service credential, in the form credentials/{name}. Supply this field when creating the service or rebinding its credential. On read, it reflects the credential’s current name.

classmethod from_dict(

value: dict,

) → Self

as_dict(

self,

) → dict

class Privilege

SELECT = 'SELECT'

READ_PRIVATE_FILES = 'READ_PRIVATE_FILES'

WRITE_PRIVATE_FILES = 'WRITE_PRIVATE_FILES'

CREATE = 'CREATE'

USAGE = 'USAGE'

USE_CATALOG = 'USE_CATALOG'

USE_SCHEMA = 'USE_SCHEMA'

CREATE_SCHEMA = 'CREATE_SCHEMA'

CREATE_VIEW = 'CREATE_VIEW'

CREATE_EXTERNAL_TABLE = 'CREATE_EXTERNAL_TABLE'

CREATE_MATERIALIZED_VIEW = 'CREATE_MATERIALIZED_VIEW'

CREATE_FUNCTION = 'CREATE_FUNCTION'

CREATE_MODEL = 'CREATE_MODEL'

CREATE_CATALOG = 'CREATE_CATALOG'

CREATE_MANAGED_STORAGE = 'CREATE_MANAGED_STORAGE'

CREATE_EXTERNAL_LOCATION = 'CREATE_EXTERNAL_LOCATION'

CREATE_STORAGE_CREDENTIAL = 'CREATE_STORAGE_CREDENTIAL'

CREATE_SERVICE_CREDENTIAL = 'CREATE_SERVICE_CREDENTIAL'

ACCESS = 'ACCESS'

CREATE_SHARE = 'CREATE_SHARE'

CREATE_RECIPIENT = 'CREATE_RECIPIENT'

CREATE_PROVIDER = 'CREATE_PROVIDER'

USE_SHARE = 'USE_SHARE'

USE_RECIPIENT = 'USE_RECIPIENT'

USE_PROVIDER = 'USE_PROVIDER'

USE_MARKETPLACE_ASSETS = 'USE_MARKETPLACE_ASSETS'

SET_SHARE_PERMISSION = 'SET_SHARE_PERMISSION'

MODIFY = 'MODIFY'

REFRESH = 'REFRESH'

EXECUTE = 'EXECUTE'

READ_FILES = 'READ_FILES'

WRITE_FILES = 'WRITE_FILES'

CREATE_TABLE = 'CREATE_TABLE'

ALL_PRIVILEGES = 'ALL_PRIVILEGES'

CREATE_CONNECTION = 'CREATE_CONNECTION'

USE_CONNECTION = 'USE_CONNECTION'

APPLY_TAG = 'APPLY_TAG'

CREATE_FOREIGN_CATALOG = 'CREATE_FOREIGN_CATALOG'

CREATE_FOREIGN_SECURABLE = 'CREATE_FOREIGN_SECURABLE'

MANAGE_ALLOWLIST = 'MANAGE_ALLOWLIST'

CREATE_VOLUME = 'CREATE_VOLUME'

CREATE_EXTERNAL_VOLUME = 'CREATE_EXTERNAL_VOLUME'

READ_VOLUME = 'READ_VOLUME'

WRITE_VOLUME = 'WRITE_VOLUME'

MANAGE = 'MANAGE'

BROWSE = 'BROWSE'

CREATE_CLEAN_ROOM = 'CREATE_CLEAN_ROOM'

MODIFY_CLEAN_ROOM = 'MODIFY_CLEAN_ROOM'

EXECUTE_CLEAN_ROOM_TASK = 'EXECUTE_CLEAN_ROOM_TASK'

EXTERNAL_USE_SCHEMA = 'EXTERNAL_USE_SCHEMA'

READ_METADATA = 'READ_METADATA'

EXTERNAL_USE_LOCATION = 'EXTERNAL_USE_LOCATION'

class PrivilegeAssignment

principal: str | None = None

The principal (user email address or group name). For deleted principals, principal is empty while principal_id is populated.

privileges: list[Privilege]

The privileges assigned to the principal.

classmethod from_dict(

value: dict,

) → Self

as_dict(

self,

) → dict

class RateLimit

A rate limit applied to service requests. Leave requests or tokens unset to impose no limit on that dimension; set a value to cap that dimension within the renewal period.

key: RateLimitRateLimitKey

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.

renewal_period: RateLimitRateLimitRenewalPeriod

Renewal period.

principal: str | None = None

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.

requests: int | None = None

Maximum requests allowed in one renewal period. Leave unset for no request limit. Set to 0 to deny all requests.

tokens: int | None = None

Maximum tokens allowed in one renewal period. Leave unset for no token limit. Set to 0 to deny all requests.

classmethod from_dict(

value: dict,

) → Self

as_dict(

self,

) → dict

class RateLimitRateLimitKey

Scope key for a rate limit.

RATE_LIMIT_KEY_USER = 'RATE_LIMIT_KEY_USER'

RATE_LIMIT_KEY_USER_GROUP = 'RATE_LIMIT_KEY_USER_GROUP'

RATE_LIMIT_KEY_SERVICE_PRINCIPAL = 'RATE_LIMIT_KEY_SERVICE_PRINCIPAL'

RATE_LIMIT_KEY_SERVICE = 'RATE_LIMIT_KEY_SERVICE'

RATE_LIMIT_KEY_USER_DEFAULT = 'RATE_LIMIT_KEY_USER_DEFAULT'

class RateLimitRateLimitRenewalPeriod

Renewal period for a rate limit.

RATE_LIMIT_RENEWAL_PERIOD_MINUTE = 'RATE_LIMIT_RENEWAL_PERIOD_MINUTE'

RATE_LIMIT_RENEWAL_PERIOD_HOUR = 'RATE_LIMIT_RENEWAL_PERIOD_HOUR'