BerriAI/litellm · error · HTTPException

Bad Request - Either 'model' or 'completion_response' must b

Error message

Bad Request - Either 'model' or 'completion_response' must be provided

What it means

Error "Bad Request - Either 'model' or 'completion_response' must be provided" thrown in BerriAI/litellm.

Source

Thrown at litellm/proxy/spend_tracking/spend_management_endpoints.py:2149

                output_cost_per_token: Final = _litellm_params.get("output_cost_per_token")
                if input_cost_per_token is not None or output_cost_per_token is not None:
                    cost_per_token = CostPerToken(
                        input_cost_per_token=input_cost_per_token,
                        output_cost_per_token=output_cost_per_token,
                    )

                _cost = completion_cost(
                    model=_litellm_model_name,
                    messages=request.messages,
                    custom_cost_per_token=cost_per_token,
                )
            else:
                _cost = completion_cost(model=request.model, messages=request.messages)
        elif request.completion_response is not None:
            _completion_response: Final = litellm.ModelResponse(**request.completion_response)
            _cost = completion_cost(completion_response=_completion_response)
        else:
            raise HTTPException(
                status_code=400,
                detail="Bad Request - Either 'model' or 'completion_response' must be provided",
            )
        return {"cost": _cost}
    except Exception as e:
        if isinstance(e, HTTPException):
            raise ProxyException(
                message=getattr(e, "detail", str(e)),
                type=getattr(e, "type", "None"),
                param=getattr(e, "param", "None"),
                code=getattr(e, "status_code", status.HTTP_400_BAD_REQUEST),
            )
        error_msg: Final = f"{e}"
        raise ProxyException(
            message=getattr(e, "message", error_msg),
            type=getattr(e, "type", "None"),
            param=getattr(e, "param", "None"),
            code=getattr(e, "status_code", 500),

View on GitHub (pinned to 77b7c6c40c)

Solutions

1. Provide either 'model' (with messages) or 'completion_response' in the spend calculation request.

When it happens

Trigger: Thrown at litellm/proxy/spend_tracking/spend_management_endpoints.py:2149 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/05ccd6fcbf77c0a7. Report an issue: GitHub.