bytedance/deer-flow · warning · PydanticCustomError

unsupported_stream_mode

unsupported_stream_mode

Error message

Unsupported stream mode(s): {modes}

What it means

Pydantic validation error (code unsupported_stream_mode) raised when the stream_mode in a run payload contains mode names DeerFlow cannot serve. The validator delegates to normalize_stream_modes(); when that raises UnsupportedStreamModeError the offending mode list is joined into the message. HTTP 422 at request validation.

Source

Thrown at backend/app/gateway/run_models.py:113

            return value
        raise PydanticCustomError(
            "unsupported_run_option",
            "Run option '{option}' is not supported by DeerFlow",
            {"option": "stream_resumable"},
        )

    @field_validator("stream_mode", mode="before")
    @classmethod
    def reject_unsupported_stream_modes(cls, value: Any) -> Any:
        if value is None:
            return value
        if not isinstance(value, str) and (not isinstance(value, list) or not all(isinstance(mode, str) for mode in value)):
            return value
        try:
            normalize_stream_modes(value)
        except UnsupportedStreamModeError as exc:
            modes = ", ".join(exc.modes)
            raise PydanticCustomError(
                "unsupported_stream_mode",
                "Unsupported stream mode(s): {modes}",
                {"modes": modes},
            ) from exc
        return value

View on GitHub (pinned to 1dd6ba1acb)

Solutions

1. Check the {modes} list in the error message — it names exactly which mode(s) to remove.
2. Use the documented DeerFlow stream modes (e.g. 'messages', 'updates', and other modes accepted by normalize_stream_modes).
3. Validate modes client-side against the accepted set before sending.

Example fix

# before
{"stream_mode": ["messages", "values"]}

# after
{"stream_mode": ["messages", "updates"]}

const SUPPORTED_MODES = new Set(['messages', 'updates']); // keep in sync with DeerFlow
const modes = Array.isArray(payload.stream_mode) ? payload.stream_mode : [payload.stream_mode];
const bad = modes.filter(m => !SUPPORTED_MODES.has(m));
if (bad.length) throw new Error(`Unsupported stream mode(s): ${bad.join(', ')}`);

const isStreamMode = (m: unknown): m is string =>
  typeof m === 'string' && SUPPORTED_MODES.has(m);

catch 422 code 'unsupported_stream_mode'; the detail lists the bad modes — remove them and retry.

When it happens

Trigger: Sending "stream_mode": "values" (or a list like ["messages","updates","values"]) where one of the modes is not in the supported set; sending arbitrary strings as stream modes.

Common situations: Clients ported from LangGraph Platform expecting its full mode set; typos in mode strings ('update' vs 'updates'); SDK defaults that include a mode DeerFlow has not implemented.

Related errors

• unsupported_run_option
• str(exc)
• Invalid agent name '{name}'. Must match ^[A-Za-z0-9-]+$ (let
• Unknown model '{model}'. Use a model name defined under `mod
• Password is too common; choose a stronger password.

AI-assisted analysis of bytedance/deer-flow@1dd6ba1acb (2026-08-14). Data as JSON: /api/errors/00ec1e9933286e04. Report an issue: GitHub.