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
Codex ran out of room in the model's context window. Start a new thread or clear earlier history before retrying. #28937
What version of the Codex App are you using (From “About Codex” dialog)?
26.611.62324
What subscription do you have?
Pro
What platform is your computer?
Darwin 25.5.0 arm64 arm
What issue are you seeing?
Regular message:
"Codex ran out of room in the model's context window. Start a new thread or clear earlier history before retrying."
What steps can reproduce the bug?
Longer session.
Feedback ID: 019eccf6-1871-7201-92d4-cf0b69975590
What is the expected behavior?
Not getting this issue at all.
Additional information
I also noticed that when this happens "Context" shows 0% used for used and 0 tokens used.
When I then run /compact it works for a bit and then it shows values there.
Thanks for reporting this problem. Until recently, Codex used a separate endpoint and server-side logic for compaction. We recently switched to a more robust approach where the compaction logic is moved locally using the same endpoint as normal turns. This eliminates "remote compaction" errors and increases the stability and reliability of compaction operations. If you see further problems with compaction, please use /feedback and open a new bug report.
Activity
github-actions commentedon Jun 18, 2026
Potential duplicates detected. Please review them and close your issue if it is a duplicate.
Powered by Codex Action
foeken commentedon Jun 21, 2026
Same issue here. Remote connection. Thread works fine on 'server' (mac app), and client (mac app), but cli remote says:
Error running remote compact task: Codex ran out of room in the model's context window. Start a new thread or clear earlier history before retrying.
etraut-openai commentedon Jun 29, 2026
Thanks for reporting this problem. Until recently, Codex used a separate endpoint and server-side logic for compaction. We recently switched to a more robust approach where the compaction logic is moved locally using the same endpoint as normal turns. This eliminates "remote compaction" errors and increases the stability and reliability of compaction operations. If you see further problems with compaction, please use
/feedbackand open a new bug report.