Pro 20x weekly quota consumption increased dramatically after reset
I’d like to report a possible issue with Pro 20x Codex usage limits after the latest weekly quota reset.
Yesterday, I used roughly $200 worth of Sol usage and my weekly quota dropped by about 20%. Before the reset, a similar amount of usage would normally consume only a little over 10%.
Today, I used only about $20 worth of Sol, yet my quota dropped another 4–5%. This suggests that my effective quota after the reset may be significantly lower than it was before.
I’ve been subscribed to Pro 20x for quite some time, and this is the first time I’ve seen my quota burn this quickly.
Could you please investigate whether the quota allocation or usage-accounting method changed after the latest reset? I’m also concerned there may have been some kind of quota rebalancing across Pro 20x accounts—for example, reducing quota on previously normal accounts while correcting accounts that had abnormally low limits.
I’m not claiming this is definitely what happened, but the change in consumption rate is large enough that I’d appreciate clarification.
Please check whether there were any recent changes to Pro 20x quota allocation or accounting, especially around weekly resets.
8 Comments
Potential duplicates detected. Please review them and close your issue if it is a duplicate.
Powered by Codex Action
Additional diagnostic information for this quota issue:
Codex Feedback ID:
01a029e9-ae5c-7650-95b1-a1f1c211930cThis feedback was submitted from the Codex session associated with the post-reset abnormal quota consumption described in this issue. Please use it to correlate the client/session diagnostics with the server-side quota allocation and usage-accounting records before and after the latest weekly reset.
Additional note about the Codex feedback diagnostics: the app generated/displayed Feedback ID
01a029e9-ae5c-7650-95b1-a1f1c211930c, but the/feedbacksubmission itself repeatedly reports that the feedback could not be uploaded. Therefore, the diagnostic payload may not have reached the backend even though an ID was shown. Please do not assume the logs are available solely from that ID. If there is another way to attach or retrieve the relevant session diagnostics, please let me know.My 5x subscription expired overnight, even though I used Terra and Luna the whole time. I don't know what they're doing, this is a complete scam.
@lxjmaster I’m seeing a very similar pattern. Before OpenAI announced/fixed the recent reset/quota issue, my Pro 20x allowance behaved completely normally. Immediately after the fix/reset, my effective quota dropped sharply and started burning much faster than before.
I can’t prove this is what happened, but the timing makes me wonder whether quota was rebalanced across accounts to correct users who previously had abnormally low limits, with some of that reduction effectively coming from accounts that had normal allocations before. In other words, it feels less like a universal fix and more like the problem may have been shifted around.
That’s why I’d really like OpenAI to clarify whether any account-level quota allocation or rebalancing logic changed during the recent fix/reset, instead of only saying that limits are working as intended.
Additional details for the affected runs:
OpenAI Support also checked the server side and confirmed the current weekly window is 43% used / 57% remaining. They additionally noted a 500-credit grant dated 2026-08-16, a current ledger balance around -156.71 credits, while
credits availableshows none/0.I have asked Support to compare my actual Pro 20x weekly entitlement / quota allocation before vs. after the latest reset, because the core issue is the abrupt change in burn rate despite the same general usage pattern: before reset, ~$200 of CCS-estimated API-equivalent Sol usage was only a little over 10% of weekly quota; after reset, ~$200 consumed ~20%, and the next ~$20 consumed another ~4–5%.
I also asked whether the anomalous credit-ledger state has any effect on the included Pro 20x weekly quota calculation, or is unrelated.
so i can use the r word happily ever after, a reset
a reset now
Starting_Somewhere_Else.mp3
New data after another weekly reset: the quota burn rate has worsened again under the same configuration.
Configuration remains:
After this latest reset:
I understand that CCS API-equivalent dollars are not a direct conversion to the included Codex weekly allowance. I am using the same estimator only as a consistent relative measurement across resets. The cross-reset pattern is now:
Within the latest weekly window, the ~$85 / 13% and ~$200 / ~30% observations are also reasonably consistent with each other, which makes this harder to explain as a one-off task-complexity fluctuation.
I have asked OpenAI Support Case #13745369 to answer the key question directly: did my server-side Pro 20x weekly entitlement / quota allocation change across these resets, yes or no? If not, please inspect the server-side accounting history across the reset boundaries and explain the sharp change in burn ratio under the same client/model/mode/reasoning configuration.