Day 3/
The big one is GPT-6 in Chat, but today is also a little celebration day with a new high of 40M active users across Codex and ChatGPT Work.
Loading a banked reset in everyone's paid accounts. See you again tomorrow!
Roundup of Day 2/
2.1/ Approve for me (auto-review) is now included and does not use usage. Can be between 2-10% of plan when used. Also better for you.
2.2/ Simplified API for builders.
2.3/ Meeting notes integrated.
2.4/ Decisions API live for builders. Will use in the app to improve the experience.
Latest from Tibo · Send feedback to @keyanzhang and @sharifshameem !
Day 5/
Composer predictions in the desktop app. Often leads to a double take with how on point they are. Included in the Pro plans without consuming usage.
Now in beta: composer predictions in Codex for Pro users.
Codex can now suggest your next message based on your conversation and how you talk to it.
One of the most loved new features we've ever tested internally. https://t.co/0u8d2a3feS
Reset events · 90 days
8 events
Banked events · 90 days
4 events
Median confirmation gap
13.6 days
Latest reset confirmation
Counts recognized events. Announcements and confirmations are grouped; discussion is excluded. Counts do not establish individual account resets.
Event calendar
Choose a date for resets, banked resets and original posts.
What I wanted to say yesterday is that we hit 25M active users and to celebrate we have now reset usage for all paid subscriptions for ChatGPT Work and Codex.
See you soon for more news from The Reset Company.
Update on rate limits in Codex.
We’ve found (a) some inefficiencies when using images in long sessions with multiple compactions (b) high p95+ usage for Computer History (c) a feature that was meant to generate conversation titles that was draining a bit more usage than intended. And we have a tiger team combing through everything and shipping fixes tomorrow. We also found a novel approach to drive efficiency up significantly that is completely unrelated and we will be working on next week.
As part of some of the fixes tomorrow, we will also do a full reset of the usage for all paid subscriptions.
See you then.
Update on rate limits in Codex. We do see that for some users the cache hit rate has been worse this week than the stable state the weeks before. This could explain that usage is draining somewhat faster for those users as hitting the cache consistently is an important component of being efficient.
We are investigating and will have an update tomorrow.
Update on rate limits in Codex.
We’ve found (a) some inefficiencies when using images in long sessions with multiple compactions (b) high p95+ usage for Computer History (c) a feature that was meant to generate conversation titles that was draining a bit more usage than intended. And we have a tiger team combing through everything and shipping fixes tomorrow. We also found a novel approach to drive efficiency up significantly that is completely unrelated and we will be working on next week.
As part of some of the fixes tomorrow, we will also do a full reset of the usage for all paid subscriptions.
See you then.
Start with the original words.“Next hour” starts at publication. Explicit clock times use their stated time zone; PST and PDT use their corresponding offsets. Missing zones use an explicitly labelled Pacific-local assumption. Estimates add a separate 1–2 hour allowance.
Missing times get a reference window.The reference is 16:30–21:30 Pacific, adjusted for daylight saving time. We count confirmation-post timestamps; these do not establish when a reset button was pressed.
Of 7 collected confirmation timestamps in 90 days, 4 fall in that window.
Source facts and estimates stay separate.Only explicit confirmation posts are source-confirmed. Six hours after an estimated window, a record may say “likely completed (estimate)”; it remains unconfirmed. Banked resets do not restore usage by themselves. Posting time is not delivery time.
Public posts are checked continuously.New posts every 1.0 hours; relevant older sources every 6.0 hours. Messages publish automatically and may be withdrawn by an operator. Check your own quota and banked balance in Codex.
Earliest collected record · Coverage is limited to accessible sources successfully read by this monitor, not a complete historical archive. Manual evidence is labelled separately.
Tibo@thsottiaux
Source post
