OpenAI combines ChatGPT and Codex limits to save tokens efficiently

OpenAI has announced a major update to its usage limits system for core artificial intelligence tools. Starting this December, subscribers of the Plus, Pro, and Business tiers will draw from a shared pool of limits when working with both ChatGPT and Codex, moving away from previously separate restrictions.

• Shared token limit for ChatGPT and Codex for Plus, Pro, and Business users starting in December.
• Resource consumption depends on the chosen model complexity and agentic operations.
• The Instant mode remains an exception and won't count toward the shared limit.
• Users can check remaining limits and reset times in the settings menu.

New Token-Based Limit System

Token consumption will now be calculated directly based on tokens — the fundamental units of information that neural networks process and generate. The volume of resources spent depends on the specific model selected and the complexity of the task, with advanced models and prolonged agentic operations consuming significantly more power.

Consequently, intensive coding work within Codex will directly reduce the available token reserve for complex queries in ChatGPT, and vice versa. This structural update is tied to the deeper integration of advanced agentic capabilities from Work mode directly into the main ChatGPT interface.

Exceptions and Resource Management

The sole exception to the new pooling rules will be the Instant mode. Standard daily conversations utilizing this fast mode will not be counted against the shared limit. According to OpenAI's internal estimates, the vast majority of users should not experience a noticeable increase in waiting times for limit replenishments.

At present, OpenAI has not yet specified the exact numerical limits for each individual subscription tier, the exact reset frequency, or whether these updates will impact Free and Go users. If limits are fully depleted, users will be able to wait for the standard reset, upgrade their subscription tier, or purchase additional credits.

Checking Usage and Practical Impact

Users will be able to check their remaining shared limit and the exact time of its next reset by navigating to the Settings section and opening the Usage subsection. This provides full transparency for managing daily workloads.

For developers, tech enthusiasts, and businesses relying on advanced AI tools, this integration requires a more strategic approach to workload management. Balancing heavy coding tasks with general analytical queries becomes essential to maintaining uninterrupted workflow efficiency.