GitHub Copilot: Gemini 3.1 Pro Now Available in JetBrains, Xcode, and Eclipse
GitHub CopilotGitHub expanded access to Google's Gemini 3.1 Pro model in GitHub Copilot, making it available in public preview across JetBrains IDEs, Xcode, and Eclipse — completing its rollout to every major development environment supported by Copilot. The model is accessible through the chat model picker in agent, ask, and edit modes for users on Copilot Enterprise, Business, Pro, and Pro+ plans. Copilot Business and Enterprise administrators must enable the Gemini 3.1 Pro policy in their organization's Copilot settings before the model appears in the picker. Gemini 3.1 Pro brings a 1 million token context window, 80.6% SWE-Bench Verified pass rate, and double the reasoning performance of its predecessor.
Sources & Mentions
5 external resources covering this update
Gemini 3.1 Pro Reaches All Major IDEs in GitHub Copilot
GitHub has expanded the availability of Google's Gemini 3.1 Pro model within GitHub Copilot, completing its rollout across the full suite of supported development environments. As of March 23, 2026, developers using JetBrains IDEs, Xcode, and Eclipse can now access Gemini 3.1 Pro through the Copilot Chat model picker — joining github.com, GitHub Mobile, Visual Studio Code, and Visual Studio, where the model was already available.
What Changed in This Update
When Gemini 3.1 Pro entered public preview in February 2026, it launched with support for GitHub's web interface and the most widely used desktop IDEs. This March update closes the gap by bringing the model to the remaining supported environments: JetBrains IDEs (IntelliJ IDEA, PyCharm, WebStorm, and others), Xcode for Apple platform developers, and Eclipse for Java and enterprise developers. Developers across all these environments can now select Gemini 3.1 Pro from the model picker in agent mode, ask mode, and edit mode within Copilot Chat.
About Gemini 3.1 Pro
Gemini 3.1 Pro is Google DeepMind's flagship coding-optimized model, released in February 2026. It delivers notable performance improvements over its predecessor:
- SWE-Bench Verified: 80.6% pass rate on real-world software engineering tasks
- ARC-AGI-2: 77.1% — double the score of Gemini 3 Pro
- Context window: 1 million tokens, enabling full codebase ingestion in a single prompt
- Output capacity: 65,536 tokens per response
Availability and Administrator Configuration
Gemini 3.1 Pro in GitHub Copilot is available in public preview to subscribers on Copilot Enterprise, Business, Pro, and Pro+ plans.
For Copilot Business and Enterprise organizations, administrators must explicitly opt in by enabling the Gemini 3.1 Pro policy in their organization's Copilot settings. Individual Pro and Pro+ users can access the model directly without additional configuration.