OpenAI Merges Chat and Work with a Unified Token Budget

OpenAI will merge Chat and Work and apply a shared token budget to its paid offerings. The ceilings for high-end plans have already been raised, while some exchanges will remain outside the quota.

Raised Ceilings and Aligned Competition

OpenAI has recently doubled the usage ceilings for its most expensive offerings, likely in anticipation of the merging of the Chat and Work modes. This change comes as the structure of ChatGPT, which offers two usage modes and 11 model parameters, faces criticism. Meanwhile, Anthropic has already combined the Chat and Cowork modes in Claude.

A Single Token Budget for Plus, Pro, and Business

Starting in December, the Plus, Pro, and Business subscriptions will share a single token budget between ChatGPT and Codex, according to an updated support page from OpenAI. OpenAI plans to merge the Chat and Work modes and is beginning to integrate the agentic features of Work into the main interface. The company had already announced its intention to bring these two modes closer together.

Consumption, Exemptions, and Options for Users

Tokens correspond to the amount of information that the models process and produce, with more elaborate operations depleting the quota more quickly. Users have the option to check the number of remaining tokens as well as the reset dates in the settings. Once the ceiling is reached, they have three options: wait until the reset, upgrade to a higher plan, or purchase additional credits. Conversations conducted with Instant models will not be deducted from the shared quota.