OpenAI Expands Custom GPT Models for Enhanced User Flexibility

OpenAI has introduced a significant upgrade to its custom GPT feature in ChatGPT, now allowing users to select from a broader range of models when creating or running their personalized AI systems.

Previously restricted to GPT-4o, users can now choose from an extensive list that includes GPT-4o, o3, o4-mini, o4-mini-high, GPT-4.5, GPT-4.1, and GPT-4.1-mini, increasing flexibility and customization options for various user needs.

This latest enhancement aims to empower users by enabling them to create tailored versions of ChatGPT that combine specific instructions, additional knowledge, and unique skill sets.

OpenAI expands custom GPT feature with new model options for personalized AI

For instance, someone working on social media SEO can save their instructions in a custom GPT, streamlining their workflow by eliminating repetitive tasks. As part of this, OpenAI has also launched a sort of app store filled with user-created custom GPTs, showcasing the diversity of applications.

However, there are limitations to how users can switch between models. If a custom GPT utilizes Actions—such as API calls, web browsing, or code execution—users are confined to a narrower selection of models.

OpenAI has indicated it is working on removing this restriction in the near future, potentially expanding the options available to developers.

This upgrade is presently exclusive to Pro, Plus, and Team subscribers, with promises of accessibility for Enterprise and EDU users forthcoming.

While it’s hailed as one of the most significant updates since the feature’s launch, it remains uncertain whether these changes will enhance user adoption significantly. Industry experts are watching closely to see how this development impacts the landscape of AI-driven productivity tools.