Frequently Receiving "ServiceUnavailable" Error

API
1

I’m currently working on a project that involves using the GPT-3.5 Turbo model from OpenAI. While integrating the model into my project, I’ve been encountering frequent error as

[ERROR] ServiceUnavailableError: The server is overloaded or not ready yet.

I was wondering if anyone else has faced a similar issue or has any insights on how to mitigate this problem. Are there any specific troubleshooting steps or workarounds that I can try to resolve the “Service Unavailable” error?

2

GPT-3.5 Turbo has been having elevated API error rates for sometime today. Might be resolved in sometime.

You can track the status here. https://status.openai.com/

P.S - Even after it says that the error has been resolved, it still takes sometime for the performance to go back to normal

1 Like