Key Info
OpenAI has rolled out a series of changes and added compute to significantly decrease the disruptive "Selected model is at capacity. Please try a different model" error in Codex. When the error still occurs, requests should now eventually continue on their own rather than being blocked.
Highlights
- A series of changes plus more compute target the "Selected model is at capacity" error.
- The error is described as significantly reduced in occurrence.
- Affected requests should now eventually resume automatically instead of blocking.
- Users are asked to report cases still seen in the wild.