OpenAI unifies Chat and Work modes under shared token quotas for subscribers

OpenAI is consolidating its Chat and Work modes into a single interface, according to an updated support page.reported This structural change merges agentic features from Work directly into the main ChatGPT environment, simplifying the user experience while altering how usage is tracked and limited.This means you will have one shared limit for all your ChatGPT and Codex use, which can run out faster if you use complex models.The consolidation of agentic features into a single interface with shared token quotas requires developers to optimise workflows to minimise unnecessary processing.

The evidence

The Decoder reported that starting in December, Plus, Pro and Business plans will share a common token budget across ChatGPT and Codex.reported Tokens measure the information a model processes and generates, meaning more complex models and tasks consume this budget faster. Users can monitor their remaining balance and reset times in Settings, and once the limit is reached, they must wait for a reset, upgrade their plan or buy extra credits.

The decision

OpenAI had previously announced plans to merge Chat and Work, and this update confirms the implementation timeline. The move folds Work’s agentic capabilities into the main interface, removing the distinction between the two operating modes. This consolidation likely responds to criticism regarding the growing complexity of ChatGPT, where users previously navigated two modes and eleven model settings that even experienced users found difficult to distinguish.

What the evidence means

While chats with Instant models do not count toward the budget, heavy usage on cheaper plans may result in lower-quality results. The Decoder noted that this shift signals the end of all-you-can-prompt pricing, as complex tasks eat through the shared token budget quickly. OpenAI recently doubled usage limits for higher-priced plans, presumably to prepare for this merger, which aims to streamline the interface but introduces a shared resource constraint across different tiers.

What builders can do with it

Developers integrating with ChatGPT and Codex must account for the new shared token quota when designing applications for Plus, Pro and Business users. Since complex models consume more tokens, builders should optimise workflows to minimise unnecessary processing or leverage Instant models for simple tasks to preserve the user’s budget. Applications should also guide users to check their remaining balance and reset times in Settings, ensuring transparency about quota consumption. Builders can design fallback mechanisms that prompt users to upgrade or purchase credits when limits are reached, rather than blocking functionality entirely.