GitHub's Copilot Fusion picks and combines AI models for you, and it's now on every Copilot plan.

At its Universe conference on Wednesday, GitHub made Copilot Fusion generally available. Instead of you choosing one model, Fusion sends work to several and can have one model review or build on another's draft (GitHub changelog).

It's available on all Copilot plans in VS Code, the GitHub Copilot app, Copilot CLI and JetBrains IDEs, with Standard, High and Max effort levels. GitHub says Fusion Max scores within two points of Claude Code running Opus 5.5 Max, at about half the cost and 40% less time per task. That's GitHub's claim (GitHub, GitHub on X). Artificial Analysis, which tested it independently, says Fusion Max mixes Opus 5.5, Sonnet 5.5, GPT-6 Astra and GPT-6 Sol. In its tests that cost $5.59 per task versus $11.42 for Claude Code on Opus 5.5 (max), and Fusion led Terminal-Bench 4.0 (Artificial Analysis on X).

There's no separate fee: you pay for each underlying model at its normal rate, and paid plans get 30% off through Nov. 30 (GitHub Docs). GitHub also put Copilot projects across repos and computer use, which asks before it controls an app, into public preview in the Copilot app (GitHub changelog). Starting today, GitHub Enterprise customers can use all Copilot features without a separate subscription and pay for the AI credits they use.

Try it: pick Fusion in VS Code, the Copilot app, CLI or JetBrains and start on Standard. Max can use more credits, because you pay each model's normal rate. GitHub's docs also warn that if Fusion throws away a draft, file edits that draft already made aren't undone automatically, so review your changes before you commit.