OpenAI Will Put ChatGPT and Codex on One Shared Token Allowance for Plus, Pro and Business in December

OpenAI is collapsing two usage meters into one. Starting in December, Plus, Pro and Business subscribers will have a single allowance that covers both ChatGPT and Codex, the company's coding tool.

The change sits in an OpenAI Help Center article titled "Understanding ChatGPT and Codex usage limits," updated October 31, 2026. The core line: "Plus, Pro, and Business plans' usage allowance will be shared across ChatGPT and Codex as we bring the agentic capabilities of Work into the main ChatGPT experience."

What OpenAI says changes

Usage will be measured in tokens, the chunks of text a model reads and writes. OpenAI says "more advanced models and complex tasks generally use more of your allowance."

One carve-out: "Instant chat will not draw from your usage." Casual chatting on the default fast model stays off the meter. Heavier work, such as agent runs and advanced-model requests, shares the pool.

Subscribers will see combined remaining usage and reset times under Settings → Usage. At the limit, OpenAI says users can wait for a reset, upgrade their plan, or buy credits for eligible features.

OpenAI also made a forecast: "We expect most users won't need to wait for their usage to reset more often than they do now." That is the company's expectation, not a commitment. It also says it is preparing carefully to avoid disrupting workflows.

Where this fits

This is not the first meter change this year. OpenAI moved to token-based pricing, measured in credits per million tokens, on April 2, 2026, replacing per-message estimates that treated a quick question and a sprawling compute-heavy request as roughly equal.

Pooling had already started on personal plans. OpenAI's credits page says that for Plus and Pro, Codex, ChatGPT Work, Excel and PowerPoint can share the same agentic usage allowance when those features are available on the plan. A second help article already says Work and Codex share the usage allowance included with a plan. December extends that pool to ChatGPT itself, minus Instant.

The change also lands days after OpenAI raised Work and Codex usage on Pro plans, according to a CellCog account of the company's recent adjustments.

How steep the token spread is

The price of a request depends heavily on the model. OpenAI's help center estimates for today's Work and Codex allowance on Plus run from 5 to 45 local messages per five-hour window with GPT-6 Astra, against 350 to 3,000 with the lighter Luna model. By Four Week MBA's arithmetic, that is roughly 67 to 70 times as many messages at the ends of the ranges.

Those are not December figures. OpenAI has not said how the shared allowance will convert tokens on any plan.

The objection from heavy users

The complaints came from people who use Codex all day, and they came fast. A post on X by @vikktorrrre on October 31 flagged the change, saying ChatGPT and Codex will share the same usage limits across Plus, Pro and Business from December, with Instant chat not counting. Another user, VraserX, wrote: "Heavy Codex usage shouldn't eat into your ChatGPT allowance." LuminaBench went further: "if they make this feel terrible Anthropic can happily take my money."

The underlying worry is simple. A long coding run now draws down the same balance as everyday work in ChatGPT.

The flip side is real too. Under separate buckets, a developer could run dry in Codex while ChatGPT allowance sat unused. One pool lets that stranded capacity cover either product. Business customers also get one balance to track and forecast instead of two.

What OpenAI has not said

OpenAI has given no exact launch date beyond "December." It has published no allowance amounts by plan. TokenPost reports that OpenAI previously told subscribers about the planned change at checkout, also without a start date or plan numbers.

Codex limits have moved repeatedly over the past year. OpenAI offered temporary doublings in late December 2025, and Crypto Briefing reports that Pro 100 and Pro 500 allowances doubled in October, with a planned cut to Pro 200 canceled. That history is why "most users won't notice" will be tested against real numbers, not reassurances.

The open question is the size of the pool. If OpenAI sets it high enough, the merge is a convenience. If not, heavy Codex users will feel it first, and the limits will be visible in Settings → Usage once the December rollout begins.