xtekky/gpt4free · error · RateLimitError

Response {response.status_code}: Gateway Timeout

Error message

Response {response.status_code}: Gateway Timeout 

What it means

Sync variant: HTTP 504 Gateway Timeout is reclassified as RateLimitError, mirroring the async path. g4f treats a timing-out gateway as saturation, so callers can apply the same backoff logic as for 429.

Source

Thrown at g4f/requests/raise_for_status.py:116

        is_html = response.headers.get("content-type", "").startswith(
            "text/html"
        ) or response.text.startswith("<!DOCTYPE")
        message = response.text
    if message is None or is_html:
        if response.status_code == 520:
            message = "Unknown error (Cloudflare)"
    if response.status_code in (429, 402):
        raise RateLimitError(f"Response {response.status_code}: {message}")
    if response.status_code == 401:
        raise MissingAuthError(f"Response {response.status_code}: {message}")
    if response.status_code == 403 and is_cloudflare(response.text):
        raise CloudflareError(f"Response {response.status_code}: Cloudflare detected")
    elif response.status_code == 403 and is_openai(response.text):
        raise MissingAuthError(f"Response {response.status_code}: OpenAI Bot detected")
    elif response.status_code == 502:
        raise ResponseStatusError(f"Response {response.status_code}: Bad Gateway")
    elif response.status_code == 504:
        raise RateLimitError(f"Response {response.status_code}: Gateway Timeout ")
    elif response.status_code == 400 and "API key not valid" in message:
        raise MissingAuthError(f"Response {response.status_code}: Invalid API key")
    else:
        raise ResponseStatusError(
            f"Response {response.status_code}: {'HTML content' if is_html else message}"
        )

View on GitHub (pinned to 973504e177)

Solutions

1. Retry with backoff (same handling as RateLimitError).
2. Lower max_tokens or split the task.
3. Enable streaming to keep the connection alive.
4. Move to a faster provider.

from g4f.errors import RateLimitError
def is_gateway_timeout(err: BaseException) -> bool:
    return isinstance(err, RateLimitError) and 'Gateway Timeout' in str(err)

from g4f.errors import RateLimitError
try:
    resp = g4f.ChatCompletion.create(model, messages)
except RateLimitError as e:
    if 'Gateway Timeout' in str(e):
        resp = g4f.ChatCompletion.create(model, messages, max_tokens=512)  # smaller, retry
    else:
        time.sleep(60)
        resp = g4f.ChatCompletion.create(model, messages)

When it happens

Trigger: Sync g4f calls where long generations exceed the provider gateway's upstream timeout, or the backend queue is saturated.

Common situations: Big max_tokens on slow providers; provider under load; non-streaming requests that idle past proxy timeout.

Understand the failure class

• Timeouts: ETIMEDOUT, deadlines, and hung requests — what actually expires when a request times out.
• HTTP status errors: handling 4xx and 5xx responses — how to handle 4xx and 5xx responses properly.

Related errors

• Response {response.status}: Gateway Timeout
• Response {response.status_code}: {message}
• Failed to chat: {response.status} {error_text}
• Failed to decode JSON from PhindAi response: {text}
• PhindAi API returned success=False: {data}

AI-assisted analysis of xtekky/gpt4free@973504e177 (2026-08-14). Data as JSON: /api/errors/13ea33620763c25a. Report an issue: GitHub.