BerriAI/litellm · error · ValueError

api_base is required for LangFlow. Set it via LANGFLOW_API_B

Error message

api_base is required for LangFlow. Set it via LANGFLOW_API_BASE env var or api_base parameter.

What it means

Unlike most LiteLLM providers, LangFlow has no default public API base, so get_complete_url raises ValueError when api_base is None. The value must come from the api_base parameter or the LANGFLOW_API_BASE environment variable; the final URL is built as {api_base}/api/v1/run/{flow_id}.

Source

Thrown at litellm/llms/langflow/chat/transformation.py:98

        flow_id: Final = (model.split("/", 1)[1] if "/" in model else model).strip()
        if not flow_id:
            raise LangFlowError(
                status_code=400,
                message="flow_id is required; use model langflow/{flow_id}",
            )
        return flow_id

    def get_complete_url(
        self,
        api_base: str | None,
        api_key: str | None,
        model: str,
        optional_params: dict,
        litellm_params: dict,
        stream: bool | None = None,
    ) -> str:
        if api_base is None:
            raise ValueError(
                "api_base is required for LangFlow. Set it via LANGFLOW_API_BASE env var or api_base parameter."
            )

        api_base = api_base.rstrip("/")
        flow_id: Final = quote(self._get_flow_id(model, optional_params), safe="")
        return f"{api_base}/api/v1/run/{flow_id}"

    def _get_last_user_message(self, messages: list[AllMessageValues]) -> str:
        """Extract the text of the last user message to use as input_value."""
        for msg in reversed(messages):
            if msg.get("role") == "user":
                content = msg.get("content", "")
                if isinstance(content, list):
                    content = convert_content_list_to_str(msg)
                if not isinstance(content, str):
                    content = str(content)
                return content

View on GitHub (pinned to 6c2dcb801b)

Solutions

Set the env var: export LANGFLOW_API_BASE="http://localhost:7880"
Or pass api_base explicitly: litellm.completion(model="langflow/x", api_base="http://localhost:7880", api_key=...)
For LiteLLM proxy, set api_base (or LITELLM_PROXY_API_BASE / environment config) on the langflow model_list entry so it is passed through litellm_params
If using .env files, confirm the runner actually loads them (docker-compose env_file, --env-file, etc.)

Example fix

# before
litellm.completion(model="langflow/my-flow", messages=msgs, api_key=lf_key)

# after
litellm.completion(
    model="langflow/my-flow",
    messages=msgs,
    api_key=lf_key,
    api_base="http://localhost:7880",
)

Defensive patterns

Strategy: validation

Validate before calling

import os

def ensure_langflow_config(api_base: str | None = None) -> str:
    base = api_base or os.getenv("LANGFLOW_API_BASE")
    if not base:
        raise ValueError("LANGFLOW_API_BASE is not set; refusing to call LangFlow")
    return base

Try / catch

try:
    litellm.completion(model="langflow/x", messages=msgs, api_base=ensure_langflow_config())
except ValueError as e:
    if "api_base is required for LangFlow" in str(e):
        # config/bootstrap error — alert ops, do not retry
        ...

Prevention

Set LANGFLOW_API_BASE in the deployment manifest next to the API key, not ad hoc in shells
Add a startup health check that pings {api_base}/api/v1/flows before serving langflow traffic
Use the same env-loading path (e.g. docker-compose env_file) as other provider variables

When it happens

Trigger: Calling completion(model="langflow/x") with neither api_base=... passed nor LANGFLOW_API_BASE exported in the process environment; the env var set only in a different shell/deployment than the one running LiteLLM (common in Docker or systemd units).

Common situations: Missing or misspelled LANGFLOW_API_BASE (e.g. LANGFLOW_API_URL); env var set in .env but the file is not loaded by the proxy/worker; local LangFlow instance at http://localhost:7880 not configured after adding the provider.

Related errors

AI-assisted analysis of BerriAI/litellm@6c2dcb801b (2026-08-15). Data as JSON: /api/errors/db221b199d3fef2a. Report an issue: GitHub.