As recorded at collection: MAI-Code-1.1-Flash priced at one-quarter of the previous model

MAI-Code-1.1-Flash is a programming model tailored to GitHub Copilot tasks. It improves code-writing performance, increases token (the unit AI uses to process text) generation speed by 25%, and reduces the number of tokens needed for the same task by 25%. It also newly supports understanding and analyzing image content.

Free users and students can use it through automatic model selection. Paid users must select MAI-Code-1.1-Flash themselves. It is rolling out gradually to GitHub Copilot-supported platforms, including VS Code, Visual Studio, JetBrains IDE, Copilot CLI, and GitHub Mobile.

The model costs one-quarter as much as the previous MAI-Code-1-Flash. It costs 0.20 dollars per 1 million input tokens, 0.02 dollars per 1 million cached input tokens (input tokens stored for reuse), and 1.20 dollars per 1 million output tokens. Annual Copilot subscribers get premium requests at 0.25 times the official rates. GitHub will stop offering the previous MAI-Code-1-Flash across all Copilot services starting September 10, 2026.