OpenAI Merges ChatGPT and Codex Usage Limits

Integration of Usage Allowances

Starting in December, OpenAI will unify the usage allowances for ChatGPT, Pro, and Business plans across both ChatGPT and Codex. This change is intended to support the integration of agentic capabilities into the standard ChatGPT interface. Under the new model, usage will be measured in tokens rather than message counts, with consumption rates varying based on the specific model selected and the complexity of the task performed. Instant chat interactions will remain exempt from this usage pool.

Users can monitor their combined remaining allowance and reset schedules through the Settings menu. If a user exhausts their limit, they must wait for a reset, upgrade their subscription, or purchase additional credits to continue accessing eligible features.

Community Concerns and Impact

The announcement has met with significant pushback from the developer community, who previously relied on the separation of ChatGPT and Codex usage to manage their workflows. Many users expressed frustration that high-intensity coding tasks will now directly impact their ability to perform routine, mundane queries.

Risk of Service Disruption

Users frequently noted that the previous decoupling of these services acted as a safety net. By separating the two, a user who hit their Codex limit could still rely on ChatGPT for general tasks. Commenters highlighted the potential for "service lockout," where heavy coding usage renders the entire platform unusable for simple web searches or conversational tasks.

"It kills me to run out of claude code usage then my much more mundane usage, queries or web searches, is also cut off. Not doing that has been a big plus of chatgpt/codex in my mind." — furyofantares

Transparency and Cost Predictability

Critics have pointed to a lack of transparency regarding how tokens are calculated and how much of the allowance is consumed by specific tasks. There is concern that the move is a cost-cutting measure designed to curb "token-burning" workarounds, where users utilized the standard chat interface to perform agentic tasks that would otherwise count against stricter usage quotas.

"Your usage will be measured in tokens, which we do not define, nor give you a way to measure them. The amount of tokens you consume will depend on things you can't control." — jwr

Competitive Landscape

Some users indicated that this change reduces the competitive advantage OpenAI held over providers like Anthropic. For researchers and developers, the ability to maintain separate usage pools was a primary reason for choosing OpenAI's Pro plans. With the consolidation of these limits, some subscribers are now reconsidering their loyalty to the platform, citing the potential for switching to open-weight models or alternative providers.

"I'm not surprised. At work, we paired devs and non-devs together, and we're sharing subscriptions, since we tend to exclusively use one or the other. It's like when Copilot was billing per query rather than per token, always bound to disappear." — lbreakjai