GitHub Copilot expands AI model support

GitHub is bringing multi-model support to GitHub Copilot, its AI-powered coding assistant, enabling developers to select from multiple generative AI models including Google Gemini 1.5 Pro, OpenAI GPT-4o, OpenAI o1-preview, OpenAI o1-mini, and Anthropic Claude 3.5 Sonnet. Multi-model support is only available in GitHub Copilot Chat initially, but will be enabled elsewhere in Copilot soon, GitHub said.

This models list was unveiled on October 29. In addition to announcing multi-model support, GitHub on the same day unveiled GitHub Spark, an AI-native tool for building fully functional web applications in natural language.

Developers using GitHub Copilot in github.com or Visual Studio Code can toggle between models during a conversation with Copilot Chat to choose the right model for the right use case. Or, they can continue to let GitHub Copilot use its powerful default, GitHub said. With the multi-model approach, GitHub is enabling developers to use an array of leading models in workflows they are accustomed to. Anthropic Claude 3.5 Sonnet is available via GitHub Copilot starting today while Google Gemini 1.5 Pro will be available in coming weeks. Access to the new OpenAI models is available now. GitHub will continue to enable developer choice in partnership with leading model providers and bring multi-model choice across more of Copilot’s surface areas and functions soon, the company said.

GitHub Spark, meanwhile, makes it easy for developers of all skill ranges to bring ideas to life by using natural language to build micro apps called a “spark,” GitHub said. Sparks are fully functional micro apps that can integrate AI features into external data sources without requiring any management of cloud resources. Using a creativity feedback loop, users start with an initial prompt using both OpenAI and Anthropic models, see live previews of their app as it is built, and save versions of each iteration so they can compare versions. A spark then can be run on the desktop, tablet, or mobile device. GitHub plans to iterate Spark to make the tool as intuitive as possible for consumers and developers of all skill ranges, the company said.

GitHub also said it has made substantial updates to GitHub Copilot for use in the Visual Studio Code editor, the GitHub Copilot Workspace, GitHub Models for building generative AI applications, and GitHub Copilot Autofix code scanning. Users of GitHub Copilot Chat in VS Code now can make edits across multiple files at the same time. With this new editing mode, available November 1, GitHub Copilot allows users to make complex changes across a variety of files within a project based on natural language prompts. GitHub Copilot Autofix, meanwhile, now includes security campaigns to help developers and security teams remediate vulnerabilities “at scale,” GitHub said, with the ability to triage as many as 1,000 alerts at a time.

Finally, GitHub announced that GitHub Copilot Extensions would become generally available in early-2025. GitHub Copilot Extensions, now in preview, allow developers to ask questions of any Copilot-integrated developer tool or service such as Atlassian Rovo, Docker, Sentry, and Stack Overflow. Additionally, the company said that code completion capabilities of GitHub Copilot now are available in a public preview of the Apple’s Xcode IDE.