Intel / article

Gemini 3.8 Flash arrives in GitHub Copilot across IDEs, CLI and cloud agent

Source: github.blog
Author: githubTop Viber
Date: Published Sep 3, 2026

Context

GitHub added Gemini 3.8 Flash, the newest Flash model from Google DeepMind, to the GitHub Copilot model picker on September 3, 2026. According to the changelog, it covers Copilot Pro, Pro+, Max, Business and Enterprise, and appears in Visual Studio Code, Visual Studio, JetBrains IDEs, Xcode, Eclipse, the Copilot CLI, the Copilot cloud and the GitHub Copilot app. The rollout is gradual, so the model may not show up for every account at once.

For organizations, the default matters more than the list. Under Copilot's default model enablement, new models switch on automatically for Business and Enterprise unless an administrator has turned off the global default or disables this model in the model policy. Until December 31, 2026, the model is billed at introductory provider pricing under usage-based billing. Google's own launch post says 3.8 Flash gets its gains partly by working harder, running extra reasoning steps and calling tools repeatedly, and that it can use more to do so, especially at higher effort levels.

GitHub's description of the model's behavior comes from its own early testing: it reports strong performance on complex terminal-based coding tasks and rigorous validation with persistent recovery from actionable failures. The changelog gives no scores or test details behind those observations.

Terms in this piece · Glossary

AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.
token — The chunk of text a model reads and writes in — roughly three-quarters of a word — and the unit AI usage is billed in.

Read the source github.blog