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
Weekly allowance reaches its limit much sooner after reset despite using the same model #45633
After my weekly Codex / ChatGPT Work allowance resets, I now reach the usage limit noticeably sooner than I did previously. I use exactly the same model before and after the change.
This report concerns the weekly allowance, rather than the five-hour window or a conversation's context capacity.
Confirmed observations
The weekly allowance has reset before the faster depletion is observed.
I reach the usage limit noticeably sooner in the new weekly window compared with previous usage.
The model used before and after is identical.
Details not yet recorded
The exact model label, affected client/version, reset date and time, whether the reset was scheduled or redeemed/purchased, reasoning effort, Fast setting, and before/after usage measurements are not yet included. Comparable task volume and context size have not been established. Please do not infer that all of these variables were unchanged or that I purchased a reset.
This is a qualitative user observation, not a measured claim of a specific multiplier or proof of a billing/accounting fault.
Investigation requested
Please advise how to submit account-specific diagnostics privately, and investigate whether:
The full expected weekly entitlement was restored.
Any entitlement or metering changes could explain the difference.
Unexpected background usage or a usage-accounting issue contributed to the faster depletion.
If abnormal deductions are confirmed, please review whether the affected allowance can be restored.
Activity
github-actions commentedon Sep 15, 2026
Potential duplicates detected. Please review them and close your issue if it is a duplicate.
Powered by Codex Action