Back to rankings

GitHub Copilot: MAI-Code-1.1-Flash Now Available

GitHub CopilotView original changelog

Microsoft's MAI-Code-1.1-Flash, the successor to its first in-house Copilot coding model, is now rolling out across GitHub Copilot with native vision support and gains in coding quality, instruction following, and tool use. GitHub prices it 73% lower than MAI-Code-1-Flash, with annual subscribers billed at a 0.25x premium request multiplier. It reaches Free and Student users automatically and is manually selectable on Pro, Pro+, Max, Business, and Enterprise across CLI, VS Code, Visual Studio, JetBrains, Eclipse, Xcode, GitHub Mobile, the cloud agent, and the Copilot app.

Key Takeaways

  • MAI-Code-1.1-Flash adds native vision support, letting the model interpret screenshots and diagrams passed into Copilot Chat.
  • The model is 73% cheaper by list price than MAI-Code-1-Flash, aimed at high-volume, lightweight coding tasks.
  • Free and Student users get it automatically through auto model selection, no manual setup required.
  • Paid-tier users (Pro, Pro+, Max, Business, Enterprise) can select it manually in addition to auto selection.
  • Availability spans ten surfaces, from Copilot CLI and the cloud agent to Visual Studio, JetBrains, Eclipse, and Xcode.
  • Business and Enterprise admins must opt in via policy — the model ships disabled by default for those plans.

A Faster, Cheaper Successor to MAI-Code-1-Flash

Microsoft's MAI-Code-1.1-Flash, the successor to its first in-house Copilot coding model, is now rolling out across GitHub Copilot. The new small-tier model adds native vision capabilities for image understanding — letting it reason over screenshots and diagrams passed into chat — and posts improvements in coding quality, instruction following, tool use, and overall performance compared with the original MAI-Code-1-Flash.

Pricing

GitHub says the model carries a list price 73% lower than MAI-Code-1-Flash, positioning it as a cost-effective option for lightweight, high-volume coding tasks. Annual Copilot subscribers are billed at a 0.25x premium request multiplier when using it.

Availability

MAI-Code-1.1-Flash is available automatically via auto model selection for Copilot Free and Student users, and can be manually selected by Copilot Pro, Pro+, Max, Business, and Enterprise subscribers. It is rolling out across Copilot CLI, the cloud agent, the GitHub Copilot app, Copilot Chat, VS Code, Visual Studio, GitHub Mobile, JetBrains IDEs, Eclipse, and Xcode. Business and Enterprise administrators must enable the model's policy manually in Copilot settings, since it ships off by default for those plans.

The launch comes paired with a deprecation notice for the model it replaces: GitHub is retiring MAI-Code-1-Flash on September 10, 2026.