Some users noted that these rate limit resets are not communicated clearly enough, so I decided to post this here.
I’m interested in feedback on whether this is useful for the community.
And, what are you going to build with these extra tokens?
3 Likes
This is not a gift of extra “tokens” unless your plan happens to be exhausted and waiting for a reset.
“I have reset the reset time, so instead of you getting a new period of use at the end of your current window (that might be the 50% remainder of your use today that you can see and plan in the Codex dashboard), and a reset back to 100% tomorrow, all users have a new maximum usage they can obtain within 5 days.”
A clever strategy to moderate computation over a work week… or to ensure plan limits get hit, needing conversion to credit purchases?
2 Likes
I’ll bring this up with the team directly and will also link the topic where this has been discussed previously. Thanks for raising this.
3 Likes
Yes, getting these X post with updates on this forum is appreciated.
2 Likes
Feedback: Reddit is getting wise to when ChatGPT Codex reset is not always a benefit:
What is just generally odd: type away at GPT-5.5-extended-thinking in ChatGPT, use code interpreter, upload your files, prompt a Pro subscription with your long context instuction and have it run for an hour with its shell tool… but if you do the same with remote software now using the same model, a different pool of metered AI model usage for some reason.
1 Like
These resets are likely opportunistic to save OpenAI money and hurt user usage overall. It messes up planning when we have time to utilize all of our usage, and ends up delaying resets that we may have wanted earlier to line up with our schedule (giving us usage during a time we don’t need). It must be deterministic or it is plainly unfair and disrespectful to paying customers time and $.
1 Like