GPT-5.4 mini General Availability

OpenAI's GPT-5.4 mini is now rolling out to GitHub Copilot users. This is the fast, lightweight version of the agentic GPT-5.4 model, designed for rapid inference and codebase exploration. Early testing shows it's OpenAI's highest-performing mini model to date, with exceptional speed-to-first-token and improved effectiveness with command-line tools like grep.

Where You Can Use It

The model is available to Copilot Pro, Pro+, Business, and Enterprise users across a comprehensive set of tools:

• Code Editors: Visual Studio Code, Visual Studio, JetBrains IDEs, Xcode, and Eclipse
• Web & Mobile: github.com, GitHub CLI, and GitHub Mobile (iOS and Android)
• Supported Modes: Chat, ask, edit, and agent modes (availability varies by IDE)

Enabling the Model

For Enterprise & Business Plans: Organization administrators must enable the GPT-5.4 mini policy in Copilot settings. Once enabled, users will see the model in the picker within their IDE.

For Individual Users with API Keys: Select "Manage Models" from the model picker, choose GPT-5.4 mini, and provide your OpenAI API key.

Important Notes

While GPT-5.4 mini is now generally available, the model currently carries a 0.33x premium request multiplier on pricing. GitHub notes this pricing is tentative and subject to change. For the best experience, users are encouraged to upgrade to the latest versions of their IDEs, as the model's prompting and parameters work optimally in recent releases.

GitHub recommends exploring their documentation on supported models to learn more and get started.