xtekky/gpt4free · error · RuntimeError

Failed to parse image URL: {chunk.decode(errors='replace')}

Error message

Failed to parse image URL: {chunk.decode(errors='replace')}

What it means

Raised in the StabilityAI SD3.5 Large Space stream when a 'complete' or 'generating' SSE event's data payload fails to parse as the expected shape: the code expects json.loads(chunk[6:]) to yield a list whose first element has a 'url' key (data[0]['url']). JSONDecodeError, KeyError (no 'url'), or TypeError (data is not a list / None handling aside) all land here, with the raw chunk embedded in the message.

Source

Thrown at g4f/Provider/hf_space/StabilityAI_SD35Large.py:91

                ) as event_response:
                    event_response.raise_for_status()
                    event = None
                    async for chunk in event_response.content:
                        if chunk.startswith(b"event: "):
                            event = chunk[7:].decode(errors="replace").strip()
                        if chunk.startswith(b"data: "):
                            if event == "error":
                                raise ResponseError(
                                    f"GPU token limit exceeded: {chunk.decode(errors='replace')}"
                                )
                            if event in ("complete", "generating"):
                                try:
                                    data = json.loads(chunk[6:])
                                    if data is None:
                                        continue
                                    url = data[0]["url"]
                                except (json.JSONDecodeError, KeyError, TypeError) as e:
                                    raise RuntimeError(
                                        f"Failed to parse image URL: {chunk.decode(errors='replace')}",
                                        e,
                                    )
                                if event == "generating":
                                    yield ImagePreview(url, prompt)
                                else:
                                    yield ImageResponse(url, prompt)
                                    break

View on GitHub (pinned to 973504e177)

Solutions

1. Inspect the chunk text in the message to see the actual payload shape; adapt the url extraction (data[0]['url']) to match.
2. If the payload is an error object, address the underlying Space error (often ZeroGPU quota — see the event: error path just above).
3. Retry — transient malformed frames during Space restarts resolve themselves.
4. If the shape permanently changed, update the parser in StabilityAI_SD35Large.py to the new Gradio output format.

try:
    ...
except RuntimeError as e:
    if 'Failed to parse image URL' in str(e):
        # inspect chunk in message; transient frames justify one retry
        ...

When it happens

Trigger: The Space returns a dict instead of a list (Gradio version change), an output entry without a url (e.g. an error/status object), a JSON payload split across SSE lines, or an HTML/HTML-fragment payload in the data frame; also fires if the event variable from a previous 'event:' line mismatches the data frame ordering.

Common situations: Gradio upgrades on the Space changing output payload shape ([{url:...}] vs {url:...}); error frames mislabeled as 'generating'; intermittent malformed frames during Space cold start.

Understand the failure class

• Parsing and encoding errors: unexpected token, malformed input — why parsers reject input and how to find the real culprit.

Related errors

• Failed to parse message: {chunk.decode(errors='replace')}
• GPU token limit exceeded: {chunk.decode(errors='replace')}
• No media files provided for image generation.
• Failed to read response: {chunk.decode(errors='replace')}
• Invalid JSON data: {rest}

AI-assisted analysis of xtekky/gpt4free@973504e177 (2026-08-14). Data as JSON: /api/errors/7652f22323bfead8. Report an issue: GitHub.