GPT-5.4 mini Now Available in GitHub Copilot

GitHub has rolled out GPT-5.4 mini, OpenAI's latest fast iteration of their agentic coding model, to GitHub Copilot users. Early testing shows this is OpenAI's highest-performing mini model to date, delivering faster time to first token and strengthened capabilities for codebase exploration and grep-style tool usage.

Where You Can Use It

The model is available across a comprehensive set of development environments:

• IDEs: Visual Studio Code, Visual Studio, JetBrains, Xcode, and Eclipse
• Web platforms: github.com and GitHub Mobile (iOS and Android)
• CLI: GitHub CLI

The model appears in all supported modes including chat, ask, edit, and agent functionality.

Availability and Access

GPT-5.4 mini is available to Copilot Pro, Pro+, Business, and Enterprise subscribers. Access methods vary by plan:

• Enterprise & Business: Organization administrators must enable the GPT-5.4 mini policy in Copilot settings; users will then see it in the model picker
• Bring Your Own Key: Users can select "Manage Models," choose GPT-5.4 mini, and provide their own OpenAI API key

Pricing and Recommendations

The model launches with a 0.33x premium request multiplier on standard Copilot pricing, though GitHub notes this pricing is tentative and subject to change. For best results, upgrade to the latest versions of your development tools, as prompting and model parameters work most effectively in current releases.