Gemini 3.6 Flash model now available in GitHub Copilot
GitHub Copilot is rolling out Google's Gemini 3.6 Flash model, designed for web and app development, coding, and agentic tasks. This update offers improved task completion and token efficiency over the previous Gemini 3.5 Flash. It is available to Copilot Pro, Pro+, Max, Business, and Enterprise users, with administrators needing to enable a preview policy for Business and Enterprise plans.
- →Gemini 3.6 Flash model enhances GitHub Copilot capabilities
- →Availability and access for Gemini 3.6 Flash
Features (1) ›
- Gemini 3.6 Flash model enhances GitHub Copilot capabilities
GitHub Copilot now includes Google's Gemini 3.6 Flash model, which is optimized for web and app development, coding, and complex agentic tasks. The model features configurable reasoning effort and supports parallel tool use, demonstrating improved performance and token efficiency in early tests.
Notes (1) ›
- Availability and access for Gemini 3.6 Flash
Gemini 3.6 Flash is gradually rolling out to Copilot Pro, Pro+, Max, Business, and Enterprise users and can be selected via the model picker in various IDEs and applications. Administrators of Copilot Enterprise and Business plans must enable the Gemini 3.6 Flash Preview policy to grant access.
https://github.blog/changelog/2026-07-21-gemini-3-6-flash-is-now-available-in-github-copilot
Related releases
- CodeQL 2.26.1 enhances analysis accuracy and framework coverage GitHub Changelog ·
- GitHub Copilot App Usage Metrics Expanded in API GitHub Changelog ·
- npm Adds Publish-Time Malware Scanning and Dual-Use Metadata GitHub Changelog ·
- Grok 4.5 reasoning model available in GitHub Copilot GitHub Changelog ·
- GitHub Actions holds workflows for approval to prevent attacks GitHub Changelog ·
- Dependabot alerts expand to cover malicious packages from OpenSSF GitHub Changelog ·