ChatGPT and Codex Will Share a Unified Token Usage Allowance Starting in December for Plus, Pro, and Business
According to a Help Center page, OpenAI will introduce a single shared usage allowance measured in tokens between ChatGPT and Codex starting in December for Plus, Pro, and Business tiers. Instant chat will not count toward the limit; neither a specific date nor tier-specific volumes have been released.
OpenAI is changing how it meters usage for its subscribers. According to a Help Center page published in late October, starting in December, the Plus, Pro, and Business plans will share a single combined usage allowance between ChatGPT and Codex, its coding agent. The company links this change to the rollout of Work agentic features into the core ChatGPT experience.
Tokens Instead of Messages
Usage will be measured in tokens, the units of text the model reads and generates. A given prompt will therefore not cost the same depending on the selected model and the task: OpenAI notes that advanced models and complex tasks will generally consume more. One notable exception: Instant chat will not count toward the limit. Combined remaining usage and reset times will be visible in settings under the Usage section. Once the limit is reached, users will need to wait, upgrade their plan, or purchase credits for eligible features.
For developers, this shift builds on an existing model: according to the Codex help page, Codex, ChatGPT Work, ChatGPT for Excel, and workspace agents already share an allowance and a credit pool on qualifying plans. The main change is bringing general ChatGPT usage itself into that pool, excluding Instant chat.
What Remains Unclear
OpenAI provides no specific date in December and no usage volumes per plan. The company says it expects most users will not have to wait for a reset more often than they do today, but this is a forecast that cannot be verified until the actual limits are disclosed.
Our take
Switching from message counts to a shared token allowance is theoretically fairer, but it makes usage less predictable: a subscriber running heavy Codex tasks could see their allowance for advanced conversational models drain rapidly, and vice versa. For developers who rely heavily on Codex under a Plus or Pro plan, the real question is what volume OpenAI will assign to each tier, which remains unknown. Until December, the Usage settings page will be the only reliable way to monitor actual consumption.
Sources
- OpenAI Help Center — Understanding ChatGPT and Codex usage limits Primary source
- OpenAI Help Center — Using Codex with your ChatGPT plan Primary source
- TokenPost — OpenAI Will Combine ChatGPT and Codex Usage Allowances in December Secondary source
A primary source comes directly from the company, lab or organization concerned. A secondary source reports the information without being its author: we flag it as such. Spotted an error? See our corrections policy.
