AI API error 500: internal errors and bounded retries
A 500 indicates an internal server error on the responding endpoint. One failed request does not establish a global AI service outage.
Quick answer
A 500 indicates an internal server error on the responding endpoint. One failed request does not establish a global AI service outage.
Provider guidance
OpenAI documents 500 as a server-side error. Claude calls its 500 response api_error and recommends exponential backoff; persistent failures should be investigated with the request ID.
Recommended application response
Verify which endpoint returned the response: a model provider, router, proxy or your own application. Save the request time, model, error type and request ID without secrets. Consult that service’s current report, including component scope and collection time.
Use a limited retry budget for a transient failure. Check whether your SDK already retried. Before repeating an interrupted operation, consider whether output, billing or downstream actions may already have occurred. A provider report does not confirm your individual request will succeed.
Related service pages
Each entry shows its own collection coverage. A linked service is not necessarily automatically collected.
Editorial review date applies to this guide. Live report collection times appear on the service pages.