OpenAI Merges ChatGPT Chat and Work Tabs, Codex Shares the Same Quota

One Entry, Two Modes: OpenAI Integrates ChatGPT and Codex

OpenAI is restructuring the ChatGPT product. According to a report from an RSS source, OpenAI will merge ChatGPT's existing "Chat" and "Work" tabs and have Codex share the same usage allocation pool with ChatGPT. This means users will no longer face a fragmented entry point when switching between different work modes, and the quota logic behind it will also be unified.

The direct result of this change is that ChatGPT is evolving from "a chat tool" into "a unified workbench." Codex, as a capability aimed at coding scenarios, is being brought into the same quota system, and both the user experience and cost perception will change accordingly.

Why Merging the Tabs Is an Inevitable Step

ChatGPT started with conversation at its core, and as its features expanded, it gradually branched out into a Work interface aimed at office and professional tasks. The cost of separate tabs is cognitive load: users have to figure out which entry point a given task belongs to, and teams have to maintain two sets of interaction logic.

Merging the tabs essentially hands "capability selection" back to the model and the context, rather than making users navigate the product first. For OpenAI, this reduces the churn caused by entry-point fragmentation and gives new features a unified place to live.

OpenAI's Codex first appeared as a series of code generation models separate from ChatGPT. Released in 2021, Codex powered early versions of GitHub Copilot, and its positioning was that of a specialized model for programming. Since then, OpenAI has gradually integrated coding capabilities into ChatGPT and launched a Work interface for professional tasks. Understanding this history helps clarify the essence of this adjustment: Codex is not a new feature, but an existing capability being reorganized at the product level, as OpenAI is consolidating previously separate model lines under a unified entry point.

Shared Quota Pool: Unifying the Ledger for Codex and ChatGPT

OpenAI has made it clear that after this adjustment, ChatGPT and Codex will deduct usage from the same allocation pool. This is worth noting because it changes users' intuition about "which feature is consuming quota."

Previously, users may have regarded coding-related calls as an independent resource; now they share the same quota with everyday conversations. For developers who use Codex heavily, quota planning needs to be reassessed; for ordinary users, the impact depends on the proportion of their coding tasks. OpenAI has not disclosed specific quota figures, but the phrase "same allocation pool" already makes the direction clear: usage management is moving toward centralization.

"Usage allocation pool" refers to the total quota the platform sets for an account, from which different features deduct based on actual calls, rather than each having its own independent limit. This design is common in SaaS products; the advantage is flexible resource scheduling and simple billing logic, and the cost is that users find it hard to intuitively judge how quickly a given type of task consumes quota. For ChatGPT, putting coding calls, which have higher inference costs, into the same pool as everyday conversations means the platform can allocate compute on demand on the backend, but users need more transparent usage prompts to avoid unexpectedly hitting the limit.

Practical Impact on Users and Developers

Simpler entry, lower decision cost. Users no longer have to remember which feature belongs where, and a task-oriented way of using the product will feel more natural.

Quota awareness needs updating. Because Codex and ChatGPT share quota, heavy coding scenarios may hit the limit faster. When teams set usage guidelines, they should incorporate coding tasks into a unified budget.

Product boundaries blur further. The lines between chat and work, conversation and coding, are deliberately being softened, consistent with the industry-wide trend of making AI assistants a "universal entry point."

Editor's Observation

This adjustment does not carry much information, but the signal is clear: OpenAI is consolidating the product interface and keeping the complexity on the backend. Merging the tabs solves the entry-point problem, and sharing the quota solves the resource-ledger problem; together, the two point to a more unified ChatGPT. The real test is whether, after unifying the quota, users can clearly perceive their own consumption rhythm—if this is not handled well, simplifying the entry point could instead create new confusion.

The information disclosed so far is limited to the tab merger and the shared quota. The specific rollout schedule and quota rules have not yet been detailed, and are worth continued attention.

One-line Summary

OpenAI merges ChatGPT's Chat and Work tabs and lets Codex share the same usage pool.

Paragraph Summary

OpenAI is restructuring the ChatGPT product by merging its existing "Chat" and "Work" tabs and having Codex share the same usage allocation pool with ChatGPT. This change reduces entry-point fragmentation, so users no longer need to figure out which interface a task belongs to, and capability selection is increasingly left to the model and context. At the same time, quota logic is moving toward centralization: coding-related calls and everyday conversations share the same quota, developers who use Codex heavily need to reassess their consumption rhythm, and how much ordinary users are affected depends on the share of coding tasks. Overall, ChatGPT is evolving from a single chat tool into a unified workbench, and product boundaries are blurring further. The specific rollout schedule and quota figures have not yet been disclosed.