BerriAI/litellm · error · BaseLLMException

error_msg (upstream response error message)

Error message

error_msg (upstream response error message)

What it means

Raised by litellm's container handler (sync path) when a container-provider HTTP endpoint returns a JSON body containing an 'error' key. LiteLLM wraps the upstream error message and status code in a BaseLLMException so callers see a uniform exception type. It mirrors whatever failure the upstream container service (e.g. a self-hosted model server) reported.

Source

Thrown at litellm/llms/custom_httpx/container_handler.py:283

                if is_multipart and "file" in kwargs:
                    files, headers = _prepare_multipart_file_upload(kwargs["file"], headers)
                    response = http_client.post(url=url, headers=headers, params=effective_params, files=files)
                else:
                    response = http_client.post(url=url, headers=headers, params=effective_params)
            else:
                raise ValueError(f"Unsupported HTTP method: {method}")

            # For binary responses, return raw content
            if returns_binary:
                return response.content

            # Check for error response
            response_json: Final = response.json()
            if "error" in response_json:
                from litellm.llms.base_llm.chat.transformation import BaseLLMException

                error_msg: Final = response_json.get("error", {}).get("message", str(response_json))
                raise BaseLLMException(
                    status_code=response.status_code,
                    message=error_msg,
                    headers=dict(response.headers),
                )

            # Parse response
            response_type: Final = RESPONSE_TYPES.get(endpoint_config["response_type"])
            if response_type:
                return response_type(**response_json)
            return response_json

        except Exception as e:
            raise e

    async def _async_handle(
        self,
        endpoint_name: str,
        container_provider_config: "BaseContainerConfig",

View on GitHub (pinned to 6c2dcb801b)

Solutions

1. Inspect the exception's message field — it is the upstream service's own error message and names the real cause
2. Check exception.status_code and headers to see if it is auth (401/403), not-found (404), or rate-limit (429) related
3. Verify the endpoint config in the container provider config: correct api_base, path, and request params for the endpoint you call
4. If 401/403, fix the credentials/token passed to the container provider
5. If 429/5xx, retry with backoff or check container service health

Example fix

# before
resp = container_handler.sync_request(endpoint_name="files", ...)

# after
from litellm.llms.base_llm.chat.transformation import BaseLLMException
try:
    resp = container_handler.sync_request(endpoint_name="files", ...)
except BaseLLMException as e:
    print(e.status_code, e.message)  # upstream's real error
    raise

from litellm.llms.base_llm.chat.transformation import BaseLLMException
try:
    resp = container_handler.sync_request(endpoint_name=name, ...)
except BaseLLMException as e:
    if e.status_code in (429, 500, 502, 503):
        raise TransientError(e.message) from e
    raise

When it happens

Trigger: Calling a container-based provider endpoint (e.g. Pinecone/container REST route) where the upstream returns JSON like {"error": {"message": "..."}} — bad request payloads, invalid IDs, upstream 4xx/5xx with JSON error bodies, or expired/invalid credentials on the container service.

Common situations: Misconfigured api_base pointing at a container service that rejects the request; deleting or fetching a resource (file, index, job) that does not exist; upstream container service rate limiting or internal errors returned as JSON.

Related errors

• Failed to transform Braintrust response: {str(e)}
• Error apply_db_fixes: {str(e)}
• Error getting prompt from Humanloop: {e.response.text}
• {str(e)}
• _body (masked response body)

AI-assisted analysis of BerriAI/litellm@6c2dcb801b (2026-08-15). Data as JSON: /api/errors/bcb755058e76c592. Report an issue: GitHub.