BerriAI/litellm · error · HTTPException

retrieval_config is required

Error message

retrieval_config is required

What it means

RAG query validation: model and messages are present, but 'retrieval_config' is missing/null/empty, so the handler has no vector store or retrieval settings to query with; rejected with 400.

Source

Thrown at litellm/proxy/rag_endpoints/endpoints.py:645

        model: Final = data.get("model")
        messages: Final = data.get("messages")
        retrieval_config: Final = data.get("retrieval_config")
        rerank: Final = data.get("rerank")
        stream: Final = data.get("stream", False)

        # Validate required fields
        if not model:
            raise HTTPException(
                status_code=400,
                detail={"error": "model is required"},
            )
        if not messages:
            raise HTTPException(
                status_code=400,
                detail={"error": "messages is required"},
            )
        if not retrieval_config:
            raise HTTPException(
                status_code=400,
                detail={"error": "retrieval_config is required"},
            )
        if not isinstance(retrieval_config, dict):
            raise HTTPException(
                status_code=400,
                detail={"error": "retrieval_config must be an object"},
            )
        if "vector_store_id" not in retrieval_config:
            raise HTTPException(
                status_code=400,
                detail={"error": "retrieval_config must contain 'vector_store_id'"},
            )
        await _authorize_nested_vector_store_ids(
            payload=retrieval_config,
            user_api_key_dict=user_api_key_dict,
        )

View on GitHub (pinned to 77b7c6c40c)

Solutions

1. Add retrieval_config to the request body with the vector store to query.

When it happens

Trigger: Thrown at litellm/proxy/rag_endpoints/endpoints.py:645 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/a1aab3afd3d155cc. Report an issue: GitHub.