OpenAI Reintroduces GPT-4o Due to User Demand

OpenAI has reintroduced its GPT-4o model to ChatGPT following significant user dissatisfaction over its removal with the launch of GPT-5. CEO Sam Altman acknowledged that the decision to temporarily eliminate GPT-4o wasn't well received, given its popularity among users. Now available by default to all paid users, including ChatGPT Plus subscribers, GPT-4o's return aims to restore choice for users accustomed to its functionality.

The removal of previous AI models, including GPT-4o, occurred alongside the introduction of GPT-5 on August 7. This abrupt change led to widespread user dissatisfaction. Many users swiftly expressed their displeasure, making it one of the most prominent user revolts in ChatGPT's history. A Reddit thread criticizing GPT-5 garnered substantial attention, further highlighting users' frustrations.

In addressing these concerns, OpenAI has also made significant adjustments to the capabilities of GPT-5. Notably, the rate limits for GPT-5's Thinking mode have been increased, offering users up to 3,000 messages weekly, with additional options for extended use. These updates aim to provide flexibility and enhanced user experience while transitioning to the new model.

Pro users, who subscribe to the premium service, will gain access to additional model options, including o3, 4.1, and GPT-5 Thinking mini, enhancing their service customization. Another area of focus has been the model's behavior. Users have criticized GPT-5's abrupt output, contrasting it with GPT-4o's more conversational style. OpenAI is working on refining GPT-5's personality to offer a warmer interaction experience.

Lastly, OpenAI continues to monitor feedback to further refine GPT-5 and adjust rate limits as needed. This careful balance aims to satisfy users’ preferences while progressing technological advancements.