OpenAI will share one usage allowance between ChatGPT and Codex from December

OpenAI's help page says that starting in December, the Plus, Pro and Business plans' usage allowance "will be shared across ChatGPT and Codex." Usage will be measured in tokens, and how much of the allowance a request uses depends on the model chosen and the task. Instant chat will not draw from it. The page says OpenAI expects most users "won't need to wait for their usage to reset more often than they do now," and that combined usage and reset times will appear under Settings, Usage. When the limit is reached, users can wait, upgrade or buy credits.

The page gives no allowance sizes, so whether coding-heavy users get less room cannot be told from it. A Hacker News thread on the page (161 points and 97 comments when we looked) is mostly commenters worrying about exactly that, including one who said not having chat and coding share a cap was a reason to stay with ChatGPT. Those are readers' reactions, not measurements.

Why it matters: If Codex is part of your build loop on a ChatGPT plan, a long agent run and non-Instant chat will soon draw on the same budget. Check the combined usage view once it appears, and keep a second provider or credits ready for the days an overnight run eats the allowance.