The news
OpenAI says that starting in December 2026, the usage allowance on ChatGPT Plus, Pro and Business plans will be shared across ChatGPT and Codex, its coding agent. It gave the news in a help centre article, "Understanding ChatGPT and Codex usage limits". OpenAI links the change to moving the agent features of ChatGPT Work into the main ChatGPT app. Today Work and Codex already share one allowance, while ordinary chat has its own limits.
The shared allowance is measured in tokens, the small pieces of text that AI models read and write. More advanced models and more complex tasks generally use more of it. Chats with the Instant model do not count. The remaining usage and reset times for both will show in ChatGPT under Settings, then Usage. At the limit, users can wait for the reset, upgrade their plan or buy credits for eligible features.
OpenAI has not given an exact start date, the size of the allowance on each plan or how often it resets. It says it expects that most users will not need to wait for a reset more often than they do now, but it has published no figures. The article does not mention the Free or Go plans.
What it means for you
- If someone on your team uses Codex and ChatGPT on the same seat, both will draw from one allowance. A long Codex session leaves less room for demanding chat work, and the other way round.
- Quick questions in Instant chat stay outside the allowance. Work with the larger models, long documents and agent tasks is what uses it up.
- OpenAI has named only Plus, Pro and Business. If your team is on another plan, check that plan's own limits page.
- The real limits are not published yet, so nobody can say how heavy users will be affected until December.
What to do now
- Open Settings, then Usage, in ChatGPT and note how much Work and Codex usage your team uses in a normal week.
- Find the people who use Codex or long agent tasks every day. They are the ones most likely to reach the shared limit first.
- On a Business plan, set monthly credit limits for your seats before December, so extra usage cannot surprise you on the bill.
- Move large automated jobs, such as processing hundreds of documents, to the OpenAI API, which is billed per token and does not use your team's ChatGPT allowance.
- Check OpenAI's help article again in December for the start date and the limits on your plan.
How Didset can help
We build AI workflows on the OpenAI API and other models, so heavy automated work runs on its own budget and your team keeps its ChatGPT allowance for daily work.
Talk to us about it
