Error Codes - UncensoredGPT

The API uses conventional HTTP status codes. Error responses have a consistent JSON body:

{
  "error": {
    "message": "Incorrect API key provided.",
    "type": "authentication_error",
    "code": "invalid_api_key"
  }
}

Status codes

Code	Meaning	Common cause and what to do
`400`	Bad Request	Malformed request body. Check the JSON and parameter types.
`401`	Unauthorized	Missing, invalid, or deleted API key. Check the `Authorization` header.
`403`	Forbidden	The key is not allowed to call this endpoint or model — for example, a key restricted to specific models or routes.
`404`	Not Found	Unknown model or resource ID. List models with `/v1/models` to confirm the name.
`422`	Unprocessable Entity	The request is syntactically valid but fails validation. The response details which field is wrong.
`429`	Too Many Requests	You hit a rate limit. Retry with exponential backoff.
`500`	Server Error	Something failed on our side. Retry with backoff; contact support if it persists.
`502` / `503`	Service Unavailable	The model is temporarily overloaded or restarting. Retry with backoff.

Timeouts and long generations

Long generations can be interrupted by client timeouts. Set a generous client timeout, and prefer streaming ("stream": true) so tokens arrive continuously and connections stay alive.

Still stuck?

Send us the request ID (when present in the error), the model, and a timestamp, and we’ll investigate: info@uncensoredgpt.ai.