OpenAI Cleans Up ChatGPT Lineup, Drops GPT-4o and GPT-4.1 Models

OpenAI may lose up to $14 billion this year, and the company is already reshaping its product strategy. Recent moves include introducing ads for Free and Go plans, and now OpenAI plans a major cleanup of its ChatGPT model lineup.

According to Neowin, OpenAI will retire several legacy AI models from the ChatGPT interface on February 13, 2026.

Multiple GPT-4-era models will leave ChatGPT

OpenAI plans to remove the following models from the consumer ChatGPT model picker: GPT-4o, GPT-4.1, GPT-4.1 mini, and OpenAI o4-mini.

These retirements add to previously announced removals, including GPT-5 Instant and GPT-5 Thinking. After February 13, users will no longer see these models as selectable options inside ChatGPT.

OpenAI will keep legacy models available through the OpenAI API for now. This decision gives developers and existing integrations more time to adapt, instead of forcing an immediate migration.

For end users, OpenAI will automatically move conversations and projects that rely on retired models to GPT-5.2 once the change takes effect.

OpenAI now concentrates development on newer GPT-5-series models. The company says these models reflect user feedback and include broader capability improvements, which makes maintaining older models less practical.

Still, the decision may frustrate part of the user base. Some users strongly prefer the tone, creativity, or behavior of models like GPT-4o, and they may not welcome a forced switch to newer versions.

This model cleanup fits into a wider restructuring effort. OpenAI wants to simplify its offerings while highlighting newer capabilities across its AI stack.

At the same time, the company reportedly seeks up to $60 billion from major tech partners. OpenAI also continues work on Prism AI Workspace, a dedicated workspace aimed at scientific research and academic writing.

Together, these moves signal a company under heavy financial pressure, but also one that wants to narrow its focus as it scales toward more advanced, enterprise-oriented AI systems.