Suddenly facing timeout errors

API
1

My api calls were all working yesterday, now today every single API response says

{
    "error": {
        "message": "Timed out generating response. Please try again with a shorter prompt or with `max_tokens` set to a lower value.",
        "type": "internal_error",
        "param": null,
        "code": "request_timeout"
    }
}

My prompt isn’t even too long, and the max tokens is set to 3000, this was never a problem before - did they update the limits or something?

2

Think the APIs might be down right now

1 Like
3

You can subscribe to the status page to learn when this issue is behind us.

1 Like
4

Happened the same to me today.

The fix was to change the API key. I used to have an old API key with a different format than the new ones. Maybe they deprecated that without warning.

5

this isn’t the only potential fix. I kept getting these timeouts with gpt 4.1/4.1-mini with the agents api. Happened twice so far. Recent fix was to change the model (I ran the agent with gpt 4 which gave a 500 error and then I made up a model [gpt-44] ran it and received a 400 error since it didn’t exist). I think switched the string to 4.1-mini and poof it worked. I kept my api key from early May. Granted, this is the 2nd time this has happened. But it CAN be fixed without changing the API key, at least temporarily. So I don’t think it’s the API key messing things up.