Due to a surge in demand for the flagship model GPT-5.6Sol in the past 48 hours, OpenAI has temporarily lifted the 5-hour usage limit for Plus, Pro, and Business subscription users to cope with the concentrated release of computing power needs. Product manager Tibo noted that the usage intensity of Codex and ChatGPT Work has been extremely high recently, and to ensure that users' programming and intelligent agent (Agent) workflows are not restricted, the platform has reset the one-time usage quota for all users.

openai

While releasing usage permissions, OpenAI is comprehensively optimizing the operational efficiency of GPT-5.6Sol. According to official information, this model upgrade will significantly reduce the actual computing power and Token consumption, thereby achieving the goal of handling more tasks under the same quota. Although specific technical details have not been disclosed, the industry generally speculates that the optimization core lies in improved Token generation efficiency and lightweight evolution of the underlying architecture.

This move is not only an emergency response by OpenAI to short-term traffic pressure but also reflects the inevitable trend of the large model industry shifting from "parameter arms race" to "practical engineering and cost reduction in operations." In the context where intelligent agent applications and automated programming are gradually becoming standard productivity tools, improvements in the operational efficiency of flagship models and the relaxation of service restrictions will further accelerate the deep penetration of AI workflows in commercial and development sectors, providing a more solid foundation for industry computing power accessibility and application explosion.