BerriAI/litellm · error · Exception

Error fetching file {element}: {response.status_code} {respo

Error message

Error fetching file {element}: {response.status_code} {response.text}

What it means

Raised while resolving file references for Gemini batch embeddings: GETting a referenced file (e.g. files/... under generativelanguage.googleapis.com) returned non-200, and the status plus body are embedded. Fires when an input element references a file that is missing or inaccessible.

Source

Thrown at litellm/llms/vertex_ai/gemini_embeddings/batch_embed_content_handler.py:73

        Args:
            input: GeminiEmbeddingInput that may contain file references
            api_key: Gemini API key
            sync_handler: HTTP client

        Returns:
            Dict mapping file name to {mime_type, uri}
        """
        input_list: Final = [input] if isinstance(input, str) else input
        resolved_files: Final[dict[str, dict[str, str]]] = {}

        for element in input_list:
            if isinstance(element, str) and _is_file_reference(element):
                url = f"https://generativelanguage.googleapis.com/v1beta/{element}"
                headers = {"x-goog-api-key": api_key}
                response = sync_handler.get(url=url, headers=headers)

                if response.status_code != 200:
                    raise Exception(f"Error fetching file {element}: {response.status_code} {response.text}")

                file_data = response.json()
                resolved_files[element] = {
                    "mime_type": file_data.get("mimeType", ""),
                    "uri": file_data.get("uri", element),
                }

        return resolved_files

    async def _async_resolve_file_references(
        self,
        input: GeminiEmbeddingInput,
        api_key: str,
        async_handler: AsyncHTTPHandler,
    ) -> dict[str, dict[str, str]]:
        """
        Async version of _resolve_file_references.

View on GitHub (pinned to 77b7c6c40c)

Solutions

1. Inspect the status/text for the file fetch failure.
2. Verify the Gemini file reference and API key permissions.

Example fix

# confirm the files/... reference exists via the Gemini files API.

When it happens

Trigger: Triggered when fetching a file element for Gemini batch embedding returns a non-success status.

Common situations: See trigger scenarios.

AI-assisted analysis of BerriAI/litellm@77b7c6c40c (2026-08-18). Data as JSON: /api/errors/b7d282ab176d6ee2. Report an issue: GitHub.