ChatGPT Codex Shared Usage Limit Starts in December

ChatGPT Codex Shared Usage Limit: One Allowance for Chat and Coding From December

The ChatGPT Codex shared usage limit starts in December, OpenAI says. Plus, Pro and Business subscribers will use one allowance for both chat and coding work. OpenAI’s Help Center says only “starting in December” and gives no exact day.

What the ChatGPT Codex shared usage limit changes

•       The allowance on Plus, Pro and Business plans will be shared across ChatGPT and Codex.

•       OpenAI linked the change to bringing the agent features of ChatGPT Work into the main ChatGPT experience.

•       Usage will be counted in tokens, the small chunks of text a model reads and writes. How much a request uses depends on the model and the task. Advanced models and complex tasks generally use more.

•       Instant chat will not draw from the allowance, so casual chats in that mode stay outside the shared pool.

What users will see

People will be able to see their combined remaining usage and reset times under Settings, then Usage. When the limit is reached, the page offers three options: wait for a reset, upgrade the plan, or buy credits to keep using eligible features.

OpenAI says it is preparing the transition carefully to avoid disrupting workflows. It expects most users will not wait for resets more often than they do now. That is OpenAI’s own forecast.

When does it start?

OpenAI has not given an exact date. Its Help Center says only “starting in December”. One report says the change takes effect on 31 December, but OpenAI has not published that date.

Recent allowance changes

Allowances for OpenAI’s agent features have been moving. In its release notes, OpenAI said on 29 October that it was raising Work and Codex usage in Pro and Business Premium plans. Pro 100 and Pro 500 plans now include twice their previous Work and Codex allowance.

The same entry said OpenAI cancelled a Pro 200 reduction announced by email on 29 September. Pro 200 subscribers who were on the lower allowance get twice their previous amount from 29 October. That brings them to the same limit as others on that plan.

Which plans are covered

OpenAI names Plus, Pro and Business. Its page does not mention Free, Go, Enterprise or Edu plans, so it is too early to say whether they are affected. The page also does not list allowance sizes for any plan, a reset schedule, or how chat and Codex use will be weighted. It does not say how the change applies to credit-based Business seats.

What the ChatGPT Codex shared usage limit means for subscribers

•       One pool, two uses. A long coding session can leave less for later chats on advanced models, and the other way round.

•       The effect depends on habits. Instant chat is outside the pool, so the impact depends on how often someone uses more capable models.

•       The size of the pool is unknown. Until OpenAI publishes plan-level budgets, it is not clear whether the combined pool will feel larger or smaller than today’s separate limits.

The upside:

•       One balance and one place to check it, under Settings, then Usage.

•       Instant chat does not count against the allowance.

•       Some Pro and Business Premium allowances were doubled on 29 October, and a planned Pro 200 cut was cancelled.

•       OpenAI expects most users will not wait for resets more often than they do now.

The concerns:

•       Heavy Codex users may use up the pool faster than before.

•       No allowance sizes have been published.

•       One report describes the move as a tightening of OpenAI’s fee structure. OpenAI’s page does not describe it that way, and the 29 October increases show allowances are still being adjusted.

•       Some paying users have objected on OpenAI’s community forum. Those posts are individual opinions and do not show the actual allowances.

The bottom line

OpenAI is merging chat and Codex usage into one token-based allowance for Plus, Pro and Business plans from December. The idea is simple, but the impact is not clear yet, because OpenAI has not shared allowance sizes or an exact date. Until it does, subscribers who rely on Codex for long sessions should check Settings, then Usage, once the combined view goes live.