QuotaReset
Source first

About QuotaReset & our sources.

QuotaReset helps developers follow public AI quota reset announcements, starting with Codex.

Where the data comes from

Our initial event and signal feed uses the Codex Resets public read-only API. We acknowledge its work collecting and classifying announcements. QuotaReset displays event metadata and original source links, rather than republishing full posts.

Tibo on X

Tibo (@thsottiaux) is the primary announcement source we highlight. His posts and replies can provide early timing or rollout updates. This version uses an intermediary feed; it does not directly ingest every X post, so new posts may appear on X before they are reflected here.

Reported, not individually verified

“Reported reset” means the upstream source classifies an event as executed. We have not independently verified its effect on every account. “Scheduled” means announced but still awaiting execution evidence. A watch is an uncertain signal.

Freshness and failures

Our server caches source requests for up to five minutes. The homepage status summary, signals summary, and recent archive are also rendered into HTML at request time, so they can be read without JavaScript. HTML responses may be cached for one minute. Open pages refresh periodically while visible. Each view shows the source timestamp. Old data is marked stale; failed requests do not become “no reset.” The feed depends on the availability and accuracy of its upstream source.

Statistics

Archive totals and average intervals come from the source. The heatmap covers the last 365 UTC dates. These statistics describe the archive and do not predict the next reset.

Corrections

For a disputed entry, open its original source link and compare the announcement with your account. Source-data corrections can be raised with Codex Resets. This site does not provide official account support.

Independent by design

QuotaReset is not affiliated with, endorsed by, or operated by OpenAI. Product names identify the services discussed.