GitHub Integrates Gemini 3.5 Flash for High-Speed Agentic Coding Workflows

GitHubGitHub

· Updated

GitHub moved Google's Gemini 3.5 Flash to general availability within the Copilot ecosystem, rolling it out across major IDEs and mobile apps. The integration targets agentic coding workflows where high cache efficiency and fast response times are more critical than the raw reasoning depth of larger flagship models.

GitHub moved Gemini 3.5 Flash to general availability across GitHub Copilot, following the official release of Google's Gemini 3.5 Flash earlier today. The model is accessible in Visual Studio Code, JetBrains, and Xcode. It delivers near-Pro quality at Flash-tier speed, optimized for tool use and high cache efficiency.
Availability
Copilot Pro, Pro+, Business, Enterprise
Supported IDEs
VS Code, Visual Studio, JetBrains, and more
Pricing
14X premium request multiplier
Minimum VS Code version
1.115.0
Minimum Visual Studio version
17.14.22 or 18.1.0

The update supports GitHub's agentic coding engine where speed is the primary bottleneck. While flagship models offer deeper reasoning, the iterative nature of an agent loop benefits from low latency. This integration mirrors OpenRouter's Gemini 3.5 Flash hosting, providing developers with high-performance alternatives for complex, multi-turn autonomous tasks.

You can select the model in Visual Studio Code version 1.115.0 or later. This rollout follows a pattern seen in Vercel's Gemini 3.5 Flash integration, which prioritizes reasoning models for agentic performance. For Copilot Business and Enterprise plans, administrators must manually enable the model policy in settings.

View the full update on github.blog
GitHub
GitHub
@github
X

📣 @GoogleAI’s Gemini 3.5 Flash is now generally available and rolling out in GitHub Copilot. Early testing shows ➡️ It has strong tool use, fast response times, and high cache efficiency ➡️ It is it well-suited for fast, iterative agentic coding workflows Try it out in @code. ⬇️ https://t.co/NvRNgYlKwm

34retweets344likes
View on X

Still wondering? A few quick answers below.

Gemini 3.5 Flash is available to users on Copilot Pro, Copilot Pro+, Copilot Business, and Copilot Enterprise tiers. While the model is now in general availability, individual developers and organization members should check their specific IDE version and settings to confirm the model is visible in their selection menu.

For Copilot Enterprise and Copilot Business plans, the model is not enabled by default. Administrators must navigate to their Copilot settings and explicitly enable the Gemini 3.5 Flash policy. Once this administrative policy is active, members of the organization will be able to select the model within their supported development environments.

You can use Gemini 3.5 Flash in Visual Studio Code version 1.115.0 or later, and Visual Studio versions 17.14.22 or 18.1.0 and later. The model is also rolling out to GitHub Mobile on iOS and Android, as well as the JetBrains, Xcode, and Eclipse development environments for iterative coding.

GitHub recommends this model for agentic workflows—where AI autonomously plans and executes multi-step tasks—because of its high cache efficiency and fast response times. These features allow for rapid, iterative loops and tool use that maintain near-Pro coding quality without the latency typically associated with larger, flagship reasoning models.

GitHub has implemented a tentative 14X premium request multiplier for Gemini 3.5 Flash usage. This pricing structure is subject to change as the rollout progresses. The model is designed to offer a balance of high-speed performance and cost-efficiency compared to the more expensive Pro-tier models available within the platform.

Every HeadsUpAI update is written based on its original source and reviewed before it's published. Read our editorial standards →

GitHub

See all AI news & updates from GitHub →

Share this update