BerriAI/litellm · error · BlackForestLabsError

Error parsing BFL response: {e}

Error message

Error parsing BFL response: {e}

What it means

After the edit job finishes polling, the handler transforms the final polled response. This error means raw_response.json() threw — the HTTP body was not valid JSON (or the response object was already consumed/closed). The exception's text is embedded in the message. The reported status_code mirrors the raw response's status code, which can be 200 even though the body was HTML/garbage.

Source

Thrown at litellm/llms/black_forest_labs/image_edit/transformation.py:297

        # BFL uses JSON, not multipart - return empty files
        return request_body, []

    def transform_image_edit_response(
        self,
        model: str,
        raw_response: httpx.Response,
        logging_obj: LiteLLMLoggingObj,
    ) -> ImageResponse:
        """
        Transform Black Forest Labs response to OpenAI-compatible ImageResponse.

        This is called with the FINAL polled response (after handler does polling).
        The response contains: {"status": "Ready", "result": {"sample": "https://..."}}
        """
        try:
            response_data: Final = raw_response.json()
        except Exception as e:
            raise BlackForestLabsError(
                status_code=raw_response.status_code,
                message=f"Error parsing BFL response: {e}",
            )

        # Get image URL from result
        image_url: Final = response_data.get("result", {}).get("sample")
        if not image_url:
            raise BlackForestLabsError(
                status_code=500,
                message="No image URL in BFL result",
            )

        # Build ImageResponse
        return ImageResponse(
            created=int(time.time()),
            data=[ImageObject(url=image_url)],
        )

View on GitHub (pinned to 6c2dcb801b)

Solutions

1. Inspect the raw body: log response.text in a retry to see what actually came back (HTML page, empty, partial JSON).
2. If a proxy is in play, bypass it for *.bfl.ai or configure it to pass JSON through unmodified.
3. Verify api_base — it should be the real BFL endpoint, not a gateway with a different response schema.
4. Retry once; transient gateway pages usually resolve.

null

null

from litellm.exceptions import APIError

try:
    resp = await litellm.aimage_edit(model=M, image=img, prompt=p)
except APIError as e:
    if "Error parsing BFL response" in str(e):
        log.warning("BFL returned non-JSON final body; retrying once")
        return await litellm.aimage_edit(model=M, image=img, prompt=p)
    raise

When it happens

Trigger: The final poll of the BFL polling_url returns a non-JSON body (HTML error page from a gateway, truncated body from a dropped connection, empty body), or a proxy/interceptor rewrites the response.

Common situations: Corporate proxies or middleboxes returning HTML auth pages; BFL gateway edge incidents serving error pages with 200; response body truncation on flaky connections; custom api_base pointing at a non-BFL endpoint that replies in a different format.

Related errors

• Error parsing initial response: {e}
• Error parsing BFL response: {e}
• Error in response object format, got None
• No image URL in BFL result
• Invalid JSON response: {raw_response.text}

AI-assisted analysis of BerriAI/litellm@6c2dcb801b (2026-08-15). Data as JSON: /api/errors/1b2738ad577d6832. Report an issue: GitHub.