You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
GPT-5 Codex is repeatedly refusing to perform large tasks. #3936
GPT-5 Codex was working through a large code migration but then failed to resolve linting and TypeScript errors repeatedly. It said the task was too big.
What is the expected behavior?
To work through the issues one by one.
What do you see instead?
No response
Additional information
No response
Reacted by Adam Mikulis, Aria, Andrew Noblet, Logan, Greg and Serg
Hi @travisjbeck, I think you are experiencing the same issue that I am in #3938. It seems to think that it's in a different environment (sandboxed and time-limited). I've noticed it since 0.39.0, so hopefully they can revert it soon.
Edit: I believe it comes from this recent change: #3874
Thanks for the bug report. We've shared the feedback with the team members who are responsible for training our codex models. We're tracking this feedback and similar reports so we can make the model more robust in future iterations.
If you see other model behavior that you'd like to report, please use the recently-added /feedback command in the CLI. That will provide us with additional details including logs and conversation text.
Activity
adammikulis commentedon Sep 19, 2025
Hi @travisjbeck, I think you are experiencing the same issue that I am in #3938. It seems to think that it's in a different environment (sandboxed and time-limited). I've noticed it since 0.39.0, so hopefully they can revert it soon.
Edit: I believe it comes from this recent change: #3874
ninjawin commentedon Sep 29, 2025
try to change model, by typing /model and select model 3 ( gpt-5-codex high).
@adammikulis @travisjbeck
etraut-openai commentedon Nov 11, 2025
Thanks for the bug report. We've shared the feedback with the team members who are responsible for training our codex models. We're tracking this feedback and similar reports so we can make the model more robust in future iterations.
If you see other model behavior that you'd like to report, please use the recently-added /feedback command in the CLI. That will provide us with additional details including logs and conversation text.