OpenAI will retire several models, including GPT-4o, from ChatGPT next month

OpenAI has revealed plans to discontinue several models from its ChatGPT platform next month, prominently including the GPT-4o model, which has garnered affection from a dedicated user base. Launched in May 2024, GPT-4o was celebrated for its engaging conversational style, particularly among a subset of paid subscribers. The decision follows a wave of criticism faced by OpenAI in August, when access to GPT-4o was temporarily halted after the introduction of its successor, GPT-5. Following user outcry, OpenAI quickly reinstated GPT-4o for its paying customers, with CEO Sam Altman assuring that ample notice would be given before any major changes to model availability. In a blog post released on Thursday, OpenAI acknowledged the frustration that some users may feel with the impending retirement of GPT-4o, stating, "This decision was not made lightly. Retiring models is challenging, but it allows us to concentrate on enhancing the models that are currently in widespread use." The company noted that just 0.1% of users engage with GPT-4o daily, while the majority have shifted to utilizing the newer GPT-5.2 model. OpenAI has been actively refining its models, enhancing aspects such as personality, customization, and creative capabilities. This is part of the rationale behind their decision to officially retire GPT-4o. In addition to GPT-4o, the company will also retire GPT-4.1, GPT-4.1 mini, and OpenAI o4-mini. Furthermore, it was previously disclosed that GPT-5 Instant and GPT-5 Thinking will also be removed from ChatGPT. Importantly, OpenAI confirmed that there will be no changes to its application programming interface (API) as these model updates take effect.