Issue: Selected model is at capacity

This is quite frustrating!

I am a Pro subscriber located in South Africa. It is now the second day that often I get “Selected model is at capacity. Please try a different model.” This is happening with SOL 5.6, SOL 6.1, Astra. One after the other in the same prompt. How is that possible? A pro subscriber value should be safeguarded about models being at capacity. I don’t think there are millions of Pro subscriber in this area of the planet. Any one else experiencing such issue as Pro subscriber?

yes, same issue. and there are similar issues posted on X. don’t know what happened to OAI currently. there should be a lot of users pivoting to Claude, therefore, such a shortage of compute is abnormal.

I hope is just a temporary issue they can fix because it is resulting in poor service.

In which country area are you guys? I am trying to understand if this is just a local issue or if it is widespread

Currentlly in Middle east and can’t use codex with Astra high and SOL 6.1 high, Pro user

I clearly paid, so why am I getting this result?

same, i doesnt saw gpt 6 like sol astra model

Hi everyone,

This could have to do with Degraded performance in Codex and ChatGPT Work.

I’m running two separate sessions for different projects right now and one of them is completely locked up with the exact same capacity error, no matter what model I switch to. Meanwhile, my other session for a different project is working perfectly fine.

It feels like specific chat threads or backend instances are getting glitched or stuck rather than the actual models being overloaded. Maybe try starting a brand-new, separate thread for your project to see if it bypasses the issue? Hopefully, they get this fixed soon!

The status page has been updated and this should be resolved.

If you are still encountering this issue please inform us here.

And thank you for raising it!

Yeah, I’m still experiencing the model capacity errors. I’m a pro 20x subscriber.

not yet,still met this issue frequently

The team has rolled out another fix, along with a set of improvements to reduce how often this issue occurs.

https://x.com/thsottiaux/status/2110750421367287996?s=20

I will close this topic and kindly ask anyone who continues to experience this issue to create a new topic. This will help us distinguish between individual reports.

Thank you for your understanding, and thanks again for raising this issue!