As a leader in the field of artificial intelligence, every technical adjustment by OpenAI always causes great ripples in the industry. Recently, OpenAI dropped a major bombshell in its official announcement: officially announcing that it will retire two classic large models, o3 and GPT-4.5, from its ChatGPT platform. The GPT-4.5, known as the "humanities genius," will be taken offline first on June 27, while the o3, focused on hard-core reasoning, will also say goodbye on August 26.

The farewell of classic models stirs feelings of nostalgia

This sudden news was hard to accept for many paid long-term users, and various social communities and discussion sections were instantly flooded with feelings of sadness and regret. Many users who were used to the "slow thinking" logic of o3 and the natural and delicate writing style of GPT-4.5 shared screenshots as mementos, and some netizens even angrily stated that this was the last reason for them to cancel their subscriptions. Although the new generation of models offer stronger overall computing power, users generally miss the unique compatibility and text warmth these two models once brought.

Product simplification aims to make way for the new king

However, it should be noted that this retirement adjustment is limited to the web and mobile app versions, while developers can still call these two models through the underlying API interface. OpenAI's decision to streamline its offerings at this moment clearly aims to optimize server resources and create an accelerated path for the upcoming new flagship models such as GPT-5.6. As the iteration speed of artificial intelligence technology approaches "monthly updates," the lifecycle of large models is rapidly shortening, and a new era of AI led by more accurate evaluations and faster iterations has already begun.