GPT-4o Code Completion Model Now Available in Public Preview for VS Code Copilot

Recently, GPT-4o Copilot has been introduced for Visual Studio Code (VS Code) users. This AI model is built on the foundation of GPT-4o mini and has undergone extensive training using more than 275,000 high-quality public repositories from over 30 widely-used programming languages. The enhanced training is expected to deliver more accurate and context-aware code suggestions, improving performance and boosting developers' productivity while aiding the coding process.

This announcement was made via a changelog published on the GitHub blog. The GPT-4o Copilot model stands out with several key enhancements. By leveraging a vast dataset of high-quality code, it provides more precise and contextually relevant code completions. Its architecture and training also enable faster and more efficient generation of code suggestions. The model supports various development projects, covering training across more than 30 programming languages.

To integrate GPT-4o Copilot into VS Code, users can access the Copilot menu in the VS Code title bar, select Configure Code Completion, and then choose Change Completion Model. Alternatively, users can open the command palette and select GitHub Copilot: Change Completion Model. Once in the model selection menu, users can choose the GPT-4o Copilot model from the available options.

For Copilot Business and Enterprise users, administrators must first enable the Copilot policy setting under Editor Preview Features on github.com to grant users access to the new model. Meanwhile, for Copilot Free users, usage of this model will count towards their monthly limit of 2,000 free completions. The model will soon be available to Copilot users across all JetBrains Integrated Development Environments (IDEs), further expanding its reach across different platforms.

In JetBrains IDEs, users can click the icon in the status bar, navigate to the settings dialog under Languages & Frameworks > GitHub Copilot, and select their preferred model from the dropdown menu.

As part of the Tech Insights 2025 Week 9 newsletter, Johan Sanneblad, CEO and co-founder of ToeknTek, commented:

GitHub Copilot is quickly catching up to the Cursor IDE, which already comes equipped with custom models. From what I've seen, Copilot only lacks two major features: prompt caching for performance and a local model for code merging. Once those two are added, I think I'll feel just as at home in Visual Studio Code + Copilot as I do in Cursor. For all Java and C# users, this is the update you've been waiting for. We finally have a code completion model that offers strong support for C++, C#, and Java.

GitHub Copilot has also made headlines for introducing next-step editing suggestions, which can predict and recommend logical edits based on the context of ongoing changes within the code. It identifies potential modifications throughout the entire file, offering suggestions for insertions, deletions, and replacements. Developers can browse these suggestions using the Tab key, streamlining the editing process and potentially saving significant time.

It is important to note that switching AI models does not affect the model used by Copilot Chat. Regardless of the selected model, data collection and usage policies remain unchanged, and settings related to enabling or disabling public code matching apply universally.

User feedback is crucial for improving and enhancing the GPT-4o Copilot model. Developers are encouraged to share their experiences to help improve the model's performance and usability, benefiting all Copilot users.