BerriAI/litellm · error · ValueError

Invalid request, No litellm managed file id or custom_llm_pr

Error message

Invalid request, No litellm managed file id or custom_llm_provider provided.

What it means

Raised in the create-fine-tuning-job handler when the request carries neither a LiteLLM-managed file id (unified_file_id) nor a custom_llm_provider: there is no routing basis for the job. Managed ids route through the router's file pipeline; an explicit provider routes through its config — with neither, the handler cannot decide where to submit the job.

Source

Thrown at litellm/proxy/fine_tuning_endpoints/endpoints.py:178

                )

            response = cast(LiteLLMFineTuningJob, await llm_router.acreate_fine_tuning_job(**data))
            response.training_file = unified_file_id
            response._hidden_params["unified_file_id"] = unified_file_id
        ## ELSE, Route based on custom_llm_provider
        elif fine_tuning_request.custom_llm_provider:
            # get configs for custom_llm_provider
            llm_provider_config: Final = get_fine_tuning_provider_config(
                custom_llm_provider=fine_tuning_request.custom_llm_provider,
            )
            # add llm_provider_config to data
            if llm_provider_config is not None:
                data.update(llm_provider_config)

            response = await litellm.acreate_fine_tuning_job(**data)

        if response is None:
            raise ValueError("Invalid request, No litellm managed file id or custom_llm_provider provided.")

        ### CALL HOOKS ### - modify outgoing data
        _response: Final = await proxy_logging_obj.post_call_success_hook(
            data=data,
            user_api_key_dict=user_api_key_dict,
            response=response,
        )
        if _response is not None and isinstance(_response, LiteLLMFineTuningJob):
            response = _response

        ### ALERTING ###
        asyncio.create_task(
            proxy_logging_obj.update_request_status(litellm_call_id=data.get("litellm_call_id", ""), status="success")
        )

        ### RESPONSE HEADERS ###
        hidden_params: Final = getattr(response, "_hidden_params", {}) or {}
        model_id: Final = hidden_params.get("model_id", None) or ""

View on GitHub (pinned to 77b7c6c40c)

Solutions

1. Provide a litellm managed file id or a custom_llm_provider in the request.

Example fix

Include custom_llm_provider=openai (or the managed file id prefix) in the fine-tuning request.

When it happens

Trigger: Thrown at litellm/proxy/fine_tuning_endpoints/endpoints.py:178 when the library encounters an invalid state.

Common situations: The fine-tuning request omitted both the managed file id and the provider.

AI-assisted analysis of BerriAI/litellm@77b7c6c40c (2026-08-18). Data as JSON: /api/errors/2f177f8283f17721. Report an issue: GitHub.