GitHub on Sept. 3, 2026, made Google’s Gemini 3.8 Flash available in GitHub Copilot, adding the model to Copilot’s picker in the Copilot app, the Copilot command-line interface, Visual Studio Code and other supported tools.
The company said in a post on X that early testing showed the model “performed strongly on complex terminal-based coding tasks” and “demonstrated rigorous validation and persistent recovery from actionable failures.” A changelog entry published the same day on GitHub’s blog repeated those results and set out which Copilot plans and products would get the model.
Google introduced Gemini 3.8 Flash a day earlier, on Sept. 2, 2026. In a company blog post, Google said the model is aimed at software engineering, agentic tasks and multi-step reasoning, and that it carries the same introductory token prices as Gemini 3.7 Flash. GitHub’s Copilot documentation lists Gemini 3.8 Flash as generally available from Google.
GitHub said Copilot Pro, Pro+, Max, Business and Enterprise users will be able to select Gemini 3.8 Flash in the model picker in Visual Studio Code, Visual Studio, Copilot CLI, the GitHub Copilot cloud agent, the GitHub Copilot app, JetBrains IDEs, Xcode and Eclipse. The company said the rollout will be gradual and that users who do not see the model yet should check again later. The changelog did not list Copilot Free among the plans that will receive the model.
Copilot Business and Enterprise administrators can allow or block the model through Copilot’s model policy settings. GitHub said it enables new models automatically unless an administrator has turned off the global default or disabled the model.
The changelog said Gemini 3.8 Flash is billed at introductory provider pricing under usage-based billing through Dec. 31, 2026. GitHub’s pricing tables list the model at $0.75 per million input tokens, $0.075 per million cached input tokens and $3.75 per million output tokens, the same promotional rates listed for Gemini 3.6 Flash and Gemini 3.7 Flash. Google’s announcement used the same introductory input and output prices and said those rates expire on Dec. 31, 2026, with $1.50 per million input tokens and $7.50 per million output tokens applying from Jan. 1, 2027.
In a separate changelog post the same day, GitHub said it will deprecate Gemini 3.5 Flash and Gemini 3.6 Flash across Copilot experiences on Oct. 2, 2026, and named Gemini 3.8 Flash as the suggested alternative for both. That notice also listed Kimi K2.7 Code and Claude Opus 4.7 for deprecation on the same date, with other suggested replacements. GitHub’s X post directed users to try Gemini 3.8 Flash in the Copilot app, CLI or Visual Studio Code.
