OpenAI announced that, beginning in December, a portion of the usage quota for ChatGPT Plus, Pro, and Business plans will be shared between ChatGPT and Codex, enabling Work agents to integrate into ChatGPT’s main interaction interface. User conversations, programming tasks, and agent operations will gradually be brought under unified quota management, measured in tokens; consumption depends on model type and task complexity, with more complex tasks consuming more; scenarios such as instant chat are also being closely monitored.
OpenAI宣布，自12月起，ChatGPT Plus、Pro和Business套餐的部分使用额度将在ChatGPT与Codex之间共享，进一步推动Work智能体能力融入ChatGPT主交互界面。这意味着用户在对话、编程及智能体任务中的资源消耗，将逐步纳入统一的额度管理体系。
根据OpenAI公告，共享额度将以Token为计量单位，实际消耗取决于所使用的模型和任务复杂度，复杂任务所需额度更高。其中，即时对话（Instant chat）不占用共享额度。OpenAI预计，这项调整不会导致大多数用户比目前更频繁地等待额度重置。
此次调整不仅涉及套餐计费规则变化，也反映出OpenAI正在整合ChatGPT与Codex等产品能力，将对话交互、代码开发和智能体执行纳入统一使用体系。通过Token计量区分不同任务的资源消耗，OpenAI有望进一步优化算力分配，同时保留轻量级即时对话的使用体验。
