OpenAI says it plans to introduce a shared ChatGPT and Codex usage allowance for Plus, Pro and Business subscribers beginning in December 2026. The company’s Help Center notice does not specify an exact start date and says Instant chat would not draw from the shared pool.
OpenAI describes the change as part of bringing the agentic capabilities of ChatGPT Work into the main ChatGPT experience. Work is its experience for longer research and deliverable tasks.
Under the proposed structure, OpenAI says usage would be measured in tokens, the units of information models process and generate. Consumption would depend on the selected model and requested task, with more advanced models and complex tasks generally using more allowance.
The company says subscribers would see their combined remaining allowance and reset times under Settings → Usage. After reaching a limit, the listed options would be to wait for a reset, upgrade or purchase credits for eligible features.
OpenAI says it is preparing for the transition to reduce workflow disruption. It expects most users will not have to wait for resets more often than before, an expectation rather than a guarantee about individual accounts.
Shared pools already cover several features
OpenAI’s Codex documentation describes an existing shared allowance and credit pool for Codex, ChatGPT Work, ChatGPT for Excel and Workspace Agents on plans that support those features. Its personal-plan credit guidance also lists ChatGPT for Word as using that allowance.
The company says consumption depends on the model, where a task runs, task complexity, context, reasoning, speed and tools. A long-running task can use substantially more allowance than a short request, so a message count alone does not reliably indicate how much work remains.
OpenAI lists Codex as included across ChatGPT plans, including Free and Go, with limits varying by plan. It lists Codex Cloud for eligible Plus, all Pro tiers, Business, Enterprise, Healthcare and Education accounts, subject to rollout and workspace settings. Cloud runs coding tasks on OpenAI-managed computers.
Depending on the plan, Work and Codex limits can apply over five-hour and weekly windows. OpenAI’s usage guidance says allowance must remain in both windows where both apply. For Plus and Business Standard, OpenAI says a full banked reset restores five-hour and weekly allowances; it can also change the next refresh time.
Its support documentation says OpenAI Support does not reset ChatGPT or Codex usage limits.
Business seats have distinct allowances
OpenAI describes ChatGPT Business as combining ChatGPT, Work and Codex through Standard and Premium seats. Premium includes more Work and Codex usage and, under the published guidance, has no five-hour usage limit. That does not make its overall included allowance unlimited.
The Business documentation provides these Standard-seat estimates for local messages within a five-hour window:
| Model | Estimated local messages |
|---|---|
| GPT-6 Sol | 15–150 |
| GPT-5.6 Luna | 250–2,000 |
OpenAI cautions that actual consumption varies by model and task, and cloud tasks can consume more allowance than local messages.
The company separately lists 15 Pro-model chat messages per month for Business Standard and 50 per week for Business Premium. Each seat’s Pro allowance is shared across GPT-6 Pro and GPT-5.6 Sol Pro, separately from Work and Codex usage.
OpenAI describes Instant chat as virtually unlimited under its Business guidance, subject to safeguards against misuse. That description does not extend to reasoning, Pro, Work, Codex or other tools. The guidance cautions that automatic reasoning can use reasoning allowance even when Instant remains selected.
October changes preceded the December plan
In its October 29, 2026 release notes, OpenAI said it doubled the Work and Codex allowance for Pro 100 and Pro 500. It also said it canceled a Pro 200 reduction announced in a September 29 email.
Pro 200 subscribers already receiving the higher allowance would retain it, the company said. Those on the lower allowance would receive twice their previous amount, bringing them to the same Pro 200 limit.
The September 29, 2026 release notes introduced Pro 500 at $500 per month. OpenAI described it as the Pro tier with the highest included usage.
