We are reseting usage for all paid users of Codex and ChatGPT Work.
Please continue reading for an update on Codex usage limits. The team has been working around the clock, going through thousands of reports and shipping fixes.
Depending on how you use Codex, you should see your usage go between 10% and 50% further than before.
We really went with a fine comb, with many uncovered small things being longstanding and here is what we found and fixed:
- Compaction. We were keeping old images during compaction, sometimes making the context large enough to trigger compaction again. After the fix, usage dropped around 10% for users making heavy use of images. Fixed.
- Memory. Background memory workers could inherit Stop hooks and keep running when the hook wouldn’t let them finish. This affected fewer than 1% of users, with the long tail being pretty bad and we saw one example thread check whether it could stop 15,000 times. Fixed.
- Goals. In some cases, a set /goal could finish and then keep going past the intended stop condition, or the model would keep retrying broken tools without stopping. We saw examples consume anywhere from 15% to 70% of a weekly allowance. Fixed.
- Automations. Some custom schedules could run more frequently than configured. Fixed.
- Subagents. Smaller models (e.g. Luna) sometimes picked more capable helpers without being explicitly asked. The same was true where the orchestrating model not running in /fast mode could request sub-agents to run /fast. Fixed.
- Computer History. The older implementation could lead to repeatedly summarizing overlapping activity. For some cases we saw it consume up to one fifth of the weekly usage per week. Fixed.
- Rolling task summaries. Ordinary turns were triggering extra background requests. These added about 1% to token usage. Small each time, but it adds up. We have disabled this.
- MCP. Some tool results could be encoded twice. We also found tool instructions getting cut off and fetched again. Fixed.
We’ve also made architectural changes to prevent these from regressing and our teams will get paged if it happens regardless. We are also working on showing you directly in the app where your usage goes so you don’t have to guess.
Goes without saying that we’re resetting usage limits and I hope you enjoy a very nice Saturday!
@jonathan_wilke Is there any reason not just spend the $ directly to the company that has the pound for pound best model?
Apart from fable I don't see why I would use another provider?
@thekitze@SnazzyLabs Sorry to hear that...
Nothing to prove really if people enjoy poop water it's on them
If you can afford better for your family it's a disservice to not provide it
Feels like people won't appreciate how big of a deal this is, Effect gives you:
- Schema
- OpenAPI client / server
- Open Telemetry
- Database drivers
- Dependency injection
- Typed errors
All with no dependencies at a time where supply chain attacks are rampant
Garbage take from people not utilizing gifts they are given properly.
WFH means you can leave your home and work remotely from a place where the weather isn't shit, it doesn't mean stay home locked in all day
WFH in the London winter made me depressed and it was a big part of why I quit my job. Every time I showered, put make up on and did my hair to go to the office was like a breath of fresh air after being underwater for a long time. We are social beings and denying ourselves the joy of interacting with other people NOT through a screen is what makes us worse
@kenwheeler@htormey I make it give me screenshots and then test the final version on device
For android dev it even served up and APK I can download and test because I was too lazy to get off the couch
@Shekswess@thsottiaux I have switched to my custom pi variant but really hoping codex fixes their shit because of the remote connection options, I really don't want to go back to TUI