BerriAI/litellm · error · BlackForestLabsError

Image generation failed: {status}

Error message

Image generation failed: {status}

What it means

Each poll reads a status field; besides 'Ready' (success) the loop treats 'Error', 'Failed', 'Content Moderated', and 'Request Moderated' as terminal failures and raises a 400 BlackForestLabsError embedding that status. This is BFL itself reporting the job died — not a LiteLLM-side or network problem.

Source

Thrown at litellm/llms/black_forest_labs/image_generation/handler.py:366

                raise BlackForestLabsError(
                    status_code=response.status_code,
                    message=f"Polling failed: {response.text}",
                )

            data = response.json()
            status = data.get("status")

            verbose_logger.debug("BFL poll status: %s", status)

            if status == "Ready":
                return response
            elif status in [
                "Error",
                "Failed",
                "Content Moderated",
                "Request Moderated",
            ]:
                raise BlackForestLabsError(
                    status_code=400,
                    message=f"Image generation failed: {status}",
                )

            time.sleep(interval)

        raise BlackForestLabsError(
            status_code=408,
            message=f"Polling timed out after {max_wait} seconds",
        )

    async def _poll_for_result_async(
        self,
        initial_response: httpx.Response,
        headers: dict,
        async_client: AsyncHTTPHandler,
        max_wait: float = DEFAULT_MAX_POLLING_TIME,
        interval: float = DEFAULT_POLLING_INTERVAL,

View on GitHub (pinned to 6c2dcb801b)

Solutions

1. Check which status string is embedded: the two 'Moderated' ones are content-policy — rephrase the prompt or change the input image.
2. 'Error'/'Failed': retry once; a subset are transient processing failures.
3. Sanitize/soften content that moderation flags, or route to a provider without equivalent policy for that content.
4. If failures cluster, check BFL status pages for incident-driven 'Error' rates.

null

null

try:
    img = litellm.image_generation(model=M, prompt=p)
except Exception as e:
    msg = str(e)
    if "Image generation failed: Error" in msg or "Image generation failed: Failed" in msg:
        return litellm.image_generation(model=M, prompt=p)  # transient processing failure
    if "Moderated" in msg:
        return litellm.image_generation(model=M, prompt=soften(p))
    raise

When it happens

Trigger: The polled job transitions to a terminal failure status: 'Error'/'Failed' from internal BFL processing, or 'Content Moderated'/'Request Moderated' when the prompt or (for edits) the image trips BFL's moderation.

Common situations: Prompts with policy-sensitive content; NSFW or violent imagery in image-edit inputs; occasionally BFL-side internal failures on hard generations; rare transient 'Error' statuses on overloaded nodes.

Related errors

• BFL error: {response_data['errors']}
• No polling_url in BFL response
• Polling failed: {response.text}
• Polling timed out after {max_wait} seconds
• Request failed: {e}

AI-assisted analysis of BerriAI/litellm@6c2dcb801b (2026-08-15). Data as JSON: /api/errors/438b60f73ea8447f. Report an issue: GitHub.