BerriAI/litellm · error · ValueError

vector_store_id is required when vertex_engine_id is not set

Error message

vector_store_id is required when vertex_engine_id is not set

What it means

Raised while building the Vertex AI Search (discoveryengine) URL: neither vertex_engine_id nor vector_store_id was supplied, so there is no data-store/engine path segment to encode. The request builder cannot construct a valid endpoint without one of them.

Source

Thrown at litellm/llms/vertex_ai/vector_stores/search_api/transformation.py:187

        vertex_location: Final = self.get_vertex_ai_location(litellm_params)
        vertex_project: Final = self.get_vertex_ai_project(litellm_params)
        collection_id: Final = litellm_params.get("vertex_collection_id") or "default_collection"
        encoded_collection_id: Final = encode_url_path_segment(collection_id, field_name="vertex_collection_id")
        base: Final = (
            f"https://discoveryengine.googleapis.com/v1/"
            f"projects/{vertex_project}/locations/{vertex_location}/"
            f"collections/{encoded_collection_id}"
        )

        engine_id: Final = litellm_params.get("vertex_engine_id")
        if engine_id:
            encoded_engine_id: Final = encode_url_path_segment(engine_id, field_name="vertex_engine_id")
            return f"{base}/engines/{encoded_engine_id}/servingConfigs/default_serving_config"

        datastore_id: Final = litellm_params.get("vector_store_id")
        if not datastore_id:
            raise ValueError("vector_store_id is required when vertex_engine_id is not set")
        encoded_datastore_id: Final = encode_url_path_segment(datastore_id, field_name="vector_store_id")
        return f"{base}/dataStores/{encoded_datastore_id}/servingConfigs/default_config"

    def transform_search_vector_store_request(
        self,
        vector_store_id: str,
        query: str | list[str],
        vector_store_search_optional_params: VectorStoreSearchOptionalRequestParams,
        api_base: str,
        litellm_logging_obj: LiteLLMLoggingObj,
        litellm_params: dict,
        extra_body: dict[str, Any] | None = None,
    ) -> tuple[str, dict[str, Any]]:
        """
        Transform a search request for the Vertex AI Search (Discovery Engine) API.

        Per-request params pass through to the engine: max_num_results maps to
        pageSize, and extra_body fields on the supported allowlist

View on GitHub (pinned to 77b7c6c40c)

Solutions

  1. Pass vector_store_id (the Vertex AI Search data store ID) in the request or configuration.
  2. Alternatively set vertex_engine_id if you intend to search an engine instead of a single data store.
Defensive patterns

Strategy: validation

When it happens

Trigger: Thrown at litellm/llms/vertex_ai/vector_stores/search_api/transformation.py:187 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/eef4ff5ca938c735. Report an issue: GitHub.