xtekky/gpt4free · warning · RuntimeError

Failed to retrieve user quota: {response.status} {error_body

Error message

Failed to retrieve user quota: {response.status} {error_body}

What it means

The quota lookup POST (get_quota) sent {'project': project_id} to the Code Assist quota endpoint and got a non-OK response; status and body are embedded. This is an informational endpoint used to report remaining quota, so failure usually mirrors a bad project ID, an auth problem, or a quota endpoint change.

Source

Thrown at g4f/Provider/needs_auth/GeminiCLI.py:1025

        url = f"{self.base_url}:retrieveUserQuota"
        headers = {
            "Authorization": f"Bearer {self.auth_manager.get_access_token()}",
            "Content-Type": "application/json",
            "User-Agent": f"GeminiCLI/1.0.0/gemini-2.5-pro ({platform.system()}; {platform.machine()})",
        }

        project_id = await self.discover_project_id()
        debug.log(f"Retrieving user quota for project: {project_id}")

        async with aiohttp.ClientSession() as session:
            async with session.post(
                url, headers=headers, json={"project": project_id}
            ) as response:
                if response.ok:
                    return await response.json()
                else:
                    error_body = await response.text()
                    raise RuntimeError(
                        f"Failed to retrieve user quota: {response.status} {error_body}"
                    )


class GeminiCLI(AsyncGeneratorProvider, ProviderModelMixin):
    label = "Google Gemini CLI"
    login_url = "https://github.com/GewoonJaap/gemini-cli-openai"

    default_model = "gemini-3-pro-preview"
    fallback_models = ["gemini-2.5-pro", "gemini-2.5-flash", "gemini-3-pro-preview"]

    working = True
    supports_message_history = True
    supports_system_message = True
    needs_auth = True
    supports_native_tools = True

    auth_manager: Optional[AuthManager] = None

View on GitHub (pinned to 973504e177)

Solutions

1. Read the embedded status and body to distinguish 401/403 (auth or project) from 5xx (transient)
2. Confirm the project id is valid and Code Assist-enabled, or unset GEMINI_PROJECT_ID to let discovery choose
3. Re-login to refresh the access token
4. Update g4f to the latest version

try:
    quota = await GeminiCLI.get_quota(api_key=creds)
except RuntimeError as e:
    if "Failed to retrieve user quota" in str(e):
        quota = None  # quota display is optional; continue generating
    else:
        raise

When it happens

Trigger: Calling GeminiCLI.get_quota() after discover_project_id when the quota endpoint returns 4xx or 5xx — for example invalid project id, expired token, or endpoint moved.

Common situations: GEMINI_PROJECT_ID points to a project without the API enabled; token expired between discovery and quota call; g4f version out of sync with the internal quota API.

Related errors

• API call failed with status {resp.status}: {text}
• Token refresh failed: {text}
• No project information found in API response.
• Could not discover project ID. Ensure authentication or set
• Token exchange failed: {error_text}

AI-assisted analysis of xtekky/gpt4free@973504e177 (2026-08-14). Data as JSON: /api/errors/6bda8889e0f332c0. Report an issue: GitHub.