ConvoShelf

ChatGPT and Codex will share one usage allowance in December: what changes

Updated November 2, 2026 · 3-minute read

Starting in December, OpenAI will measure ChatGPT and Codex usage on Plus, Pro and Business plans from one shared allowance counted in tokens, so heavy coding in Codex and long chats in ChatGPT will draw from the same pool. Instant chat won't count against it, and you'll see your combined remaining usage and reset times in Settings, under Usage.

Disclosure: ConvoShelf is our product. Everything about the usage change comes from OpenAI's Help Center article on ChatGPT and Codex usage limits, which we read on November 2, 2026, about a day after OpenAI updated it.

What OpenAI says is changing

What OpenAI hasn't said yet

The article gives no allowance sizes for any plan and no exact start date beyond "December." It doesn't mention Free, Go or Enterprise. OpenAI says it expects most users "won't need to wait for their usage to reset more often than they do now," but it doesn't publish numbers behind that. We'll update this page when OpenAI does. Until then, treat any specific figure you see quoted elsewhere with caution.

What to do before December

  1. Know which model you default to. Since usage depends on the model, a heavier default model uses the shared pool faster. Use Instant for quick questions if it's good enough, because it doesn't count.
  2. Check Settings, then Usage, once it appears. That's where OpenAI says the combined balance and reset times will show.
  3. Keep copies of chats you rely on. If you hit the limit in the middle of a project, you can still read and search what you already have. OpenAI's own export is in Settings, under Data controls; we cover the steps in our page on exporting ChatGPT chats.

Where ConvoShelf fits

ConvoShelf doesn't change your ChatGPT or Codex usage, and it can't show your remaining allowance. What it does is keep the chats you open readable and searchable in your browser, so a usage limit doesn't stop you from finding an answer you already got. We checked the code of version 1.0.2 on October 28, 2026, and the Chrome Web Store still lists 1.0.2 as the current version on November 2.

To sort what you keep, see how to organize ChatGPT chats into folders and how to search ChatGPT and Claude chats in one place.

Add ConvoShelf to Chrome Free · no account

FAQ

When does the shared ChatGPT and Codex limit start?

OpenAI's Help Center says it starts in December. As of November 2, 2026, the article doesn't give an exact date.

Which plans are affected?

OpenAI names Plus, Pro and Business. The article doesn't mention Free, Go or Enterprise.

Does chatting in ChatGPT use up my Codex allowance?

Under the new system, usage in both ChatGPT and Codex draws from one allowance measured in tokens. OpenAI says Instant chat will not draw from it.

How much usage will I get?

OpenAI hasn't published allowance sizes in this article. It says how much you use depends on the model and the task, and that it expects most users won't need to wait for resets more often than they do now.

What happens if I hit the limit?

OpenAI says you can wait for it to reset, upgrade your plan, or buy credits to keep using eligible features. Combined remaining usage and reset times will show in Settings, under Usage.

Sources

Related