BerriAI/litellm · error · BaseLLMException

Invalid response format: missing 'predictions' field

Error message

Invalid response format: missing 'predictions' field

What it means

Guard in _unwrap_predictions_response: the Vertex Gemma response JSON lacks the 'predictions' wrapper that should contain the OpenAI-compatible completion, so there is nothing to unwrap and the parent transform cannot proceed.

Source

Thrown at litellm/llms/vertex_ai/vertex_gemma_models/transformation.py:124

                {
                    "@requestFormat": "chatCompletions",
                    **openai_request,
                }
            ]
        }

    def _unwrap_predictions_response(
        self,
        response_json: dict[str, Any],
    ) -> dict[str, Any]:
        """
        Unwrap the Vertex Gemma predictions format to OpenAI format.

        Vertex Gemma wraps the OpenAI-compatible response in a 'predictions' field.
        This method extracts it so the parent class can process it normally.
        """
        if "predictions" not in response_json:
            raise BaseLLMException(
                status_code=422,
                message="Invalid response format: missing 'predictions' field",
            )

        return response_json["predictions"]

    @staticmethod
    def _sync_post(
        client: HTTPHandler | httpx.Client | None,
        api_base: str,
        headers: dict[str, str],  # mutable-ok: forwarded to post(headers: dict | None)
        request_data: dict[str, Any],  # mutable-ok: forwarded to post(json: dict | ...)
        timeout: float | httpx.Timeout | None,
    ) -> httpx.Response:
        if isinstance(client, HTTPHandler):
            return client.post(
                url=api_base,
                headers=headers,

View on GitHub (pinned to 77b7c6c40c)

Solutions

1. Inspect the raw response; missing 'predictions' usually means the endpoint returned an error payload.
2. Verify the request matches the Gemma endpoint's expected schema (instances/parameters) and retry.

When it happens

Trigger: Thrown at litellm/llms/vertex_ai/vertex_gemma_models/transformation.py:124 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/f71d42a6889c9330. Report an issue: GitHub.