BerriAI/litellm · error · HTTPException

Watsonx passthrough config not found

Error message

Watsonx passthrough config not found

What it means

Raised when ProviderConfigManager.get_provider_passthrough_config returns None for WATSONX — the Watsonx IAM token management and version-parameter injection needed by the /watsonx/* endpoints is unavailable because no provider config was registered.

Source

Thrown at litellm/proxy/pass_through_endpoints/llm_passthrough_endpoints.py:2439

    """
    Watsonx pass-through endpoint.
    Allows using Watsonx APIs with automatic IAM token management and version parameter injection.

    Example:
        POST /watsonx/ml/v1/text/tokenization
        POST /watsonx/ml/v1/text/generation
    """
    # Direct passthrough with WatsonxPassthroughConfig
    from litellm.types.utils import LlmProviders
    from litellm.utils import ProviderConfigManager

    provider_config: Final = ProviderConfigManager.get_provider_passthrough_config(
        provider=LlmProviders.WATSONX,
        model="",
    )

    if provider_config is None:
        raise HTTPException(status_code=404, detail="Watsonx passthrough config not found")

    # Get complete URL with version parameter
    complete_url, _ = provider_config.get_complete_url(
        api_base=None,
        api_key=None,
        model="",
        endpoint=endpoint,
        request_query_params=None,
        litellm_params={},
    )

    # Get auth headers with IAM token
    auth_headers: Final = provider_config.validate_environment(
        headers={},
        model="",
        messages=[],
        optional_params={},
        litellm_params={},

View on GitHub (pinned to 77b7c6c40c)

Solutions

1. Configure the Watsonx pass-through settings (WX_URL / watsonx credentials) in config.yaml or the environment.
2. Verify the Watsonx deployment is registered before calling the pass-through route.

When it happens

Trigger: Thrown at litellm/proxy/pass_through_endpoints/llm_passthrough_endpoints.py:2439 when the library encounters an invalid state.

Common situations: See trigger scenarios.

AI-assisted analysis of BerriAI/litellm@77b7c6c40c (2026-08-18). Data as JSON: /api/errors/f4ba7122818efd70. Report an issue: GitHub.