OpenAI merges Chat and Work in ChatGPT with one shared token budget from December
OpenAI is folding ChatGPT’s separate Chat and Work experiences into one interface and reshaping how paid users hit limits. According to reporting tied to an updated OpenAI support page, Plus, Pro, and Business subscribers will from December share one token budget across ChatGPT and Codex as Work’s agentic features join the main app. Coverage describes heavier models as consuming more of that pool while simple instant chats do not count toward it, eroding the old all-you-can-prompt rhythm for many tasks. Commentators in The Decoder warn the change may nudge users toward lighter models and affect perceived quality. Teams mixing coding agents and daily chat should model usage against a single cap, though the public excerpts do not spell out exact metering rules or every plan boundary.
OpenAI merges Chat and Work in ChatGPT with one shared token budget from December
ChatGPT is merging Chat and Work, changing how usage limits work. Starting in December, Plus, Pro, and Business plans will share a single token budget across ChatGPT and Codex, according to an updated OpenAI support page, as OpenAI folds Work's agentic features into the main interface.
Key takeaway
December’s unified ChatGPT and Codex token pool means Plus, Pro, and Business usage must be planned as one meter, not separate chat and work allowances.
What happened
The Decoder reports that OpenAI is combining Chat and Work into a single ChatGPT interface and that Plus, Pro, and Business plans will share a common token-based quota for ChatGPT and Codex starting in December, citing an updated OpenAI support page as OpenAI folds Work’s agentic features into the main experience.
Reporting in the packet states that more complex models consume more of the shared quota while simple instant chats do not count toward it, and The Decoder argues that in practice this shift is likely to reduce the feel of unlimited prompting and may contribute to a decline in perceived ChatGPT quality for heavy users.
Evidence
Plus, Pro, and Business plans will share a common token-based quota for ChatGPT and Codex from December.
The Decoder · attributed
Starting in December, Plus, Pro, and Business plans will share a common, token-based quota for ChatGPT and Codex.
OpenAI support page documents the merger and shared token budget as Work features move into the main interface.
The Decoder · attributed
Starting in December, Plus, Pro, and Business plans will share a single token budget across ChatGPT and Codex, according to an updated OpenAI support page, as OpenAI folds Work's agentic features into the main interface.
Heavier model use draws down the shared quota more than simple instant chats.
The Decoder · attributed
More complex models consume more of this quota, while simple instant chats do not count toward it.
The Decoder expects the quota change may affect perceived product quality.
The Decoder · attributed
In practice, this is likely to result in a decline in ChatGPT’s quality
Why it matters
A single cross-product token cap forces builders and operators to trade off Codex agent workloads against everyday chat under one commercial limit, which can change model choice, automation depth, and cost forecasting.
Limits and uncertainties
The packet does not quote exact token limits, reset intervals, or regional exceptions from the OpenAI support page.
The Decoder’s quality impact assessment is analytical rather than a confirmed OpenAI policy outcome.
Practical implications
Teams on Plus, Pro, or Business should track combined ChatGPT and Codex consumption before December instead of assuming separate chat and work headroom.
Product owners should decide default model tiers for agentic flows knowing heavier models consume more of the shared pool.
What to watch
OpenAI support documentation and December rollout messaging on how instant chats are excluded from the quota.
User and developer reports on throttling, model downgrades, or Codex limits after the merge goes live.
