xtekky/gpt4free · error · RuntimeError

\n".join([error.get("message") for error in json_line.get("e

Error message

\n".join([error.get("message") for error in json_line.get("errors")]

What it means

While parsing MetaAI's newline-delimited GraphQL stream, a JSON line contained a top-level 'errors' array. g4f joins every error's 'message' field with newlines into a RuntimeError, forwarding the GraphQL error details (as written, the f-string is missing braces so the literal expression is shown — the effective message is the joined error text).

Source

Thrown at g4f/Provider/needs_auth/MetaAI.py:142

                    "__relay_internal__pv__WebPixelRatiorelayprovider": 1,
                }
            ),
            "server_timestamps": "true",
            "doc_id": "7783822248314888",
        }
        async with self.session.post(url, headers=headers, data=payload) as response:
            await raise_for_status(response, "Fetch response failed")
            last_snippet_len = 0
            fetch_id = None
            async for line in response.content:
                if b"<h1>Something Went Wrong</h1>" in line:
                    raise ResponseError("Response: Something Went Wrong")
                try:
                    json_line = json.loads(line)
                except json.JSONDecodeError:
                    continue
                if json_line.get("errors"):
                    raise RuntimeError(
                        "\n".join(
                            [error.get("message") for error in json_line.get("errors")]
                        )
                    )
                bot_response_message = (
                    json_line.get("data", {})
                    .get("node", {})
                    .get("bot_response_message", {})
                )
                streaming_state = bot_response_message.get("streaming_state")
                fetch_id = bot_response_message.get("fetch_id") or fetch_id
                if streaming_state in ("STREAMING", "OVERALL_DONE"):
                    imagine_card = bot_response_message.get("imagine_card")
                    if imagine_card is not None:
                        imagine_session = imagine_card.get("session")
                        if imagine_session is not None:
                            imagine_medias = (
                                imagine_session.get("media_sets", {})

View on GitHub (pinned to 973504e177)

Solutions

1. Read the joined message(s): they name the exact GraphQL failure (auth vs. schema vs. policy) and dictate the next step
2. Refresh the session: recreate MetaAI / await update_cookies() to renew tokens and cookies
3. Use a residential proxy or different IP if messages mention access/permission
4. Update g4f so the current doc_id and payload are used

Example fix

// before
resp = await meta.prompt('...')  # RuntimeError with GraphQL errors

// after
try:
    resp = await meta.prompt('...')
except RuntimeError as e:
    logging.warning('MetaAI graphql error: %s', e)
    await meta.update_cookies()  # refresh and retry once
    resp = await meta.prompt('...')

try:
    resp = await meta.prompt(msg)
except RuntimeError as e:
    msg_text = str(e)
    if 'Something Went Wrong' in msg_text or 'errors' in msg_text.lower():
        await meta.update_cookies()
        resp = await meta.prompt(msg)
    else:
        raise

When it happens

Trigger: Any streamed line from the Meta AI completion endpoint with json_line['errors'] non-empty — e.g. invalid session tokens, expired access attempt, blocked prompt, or GraphQL schema/doc_id validation failures.

Common situations: Stale DTSG/LSD tokens after long-lived sessions; flagged accounts or bot-suspected IPs; prompts that trip Meta moderation; g4f version lagging a doc_id change making the query invalid.

Related errors

• Response: Something Went Wrong
• {r.status} {response body text}
• No audio chunks
• Install or update "curl_cffi" package | pip install -U curl_
• LMArena: {json.loads(line[3:])}

AI-assisted analysis of xtekky/gpt4free@973504e177 (2026-08-14). Data as JSON: /api/errors/d135de5f1dff67e7. Report an issue: GitHub.