At its Universe conference, GitHub announced on October 28, 2026, the general availability of Copilot Fusion. The core idea is that developers no longer have to manually pick a model for each task: when you select Copilot Fusion in the model picker, the system decides which model or combination of models to use and how they should work together.
Fusion combines the multi-model orchestration of Project HydraFusion with Copilot automatic model selection, aiming for a balance between performance and cost. It offers three effort levels – Standard, High, and Max – and three execution patterns: single (one model), cascade (one model drafts and another verifies or scales the work), and critique (one model drafts, another reviews, and the first one fixes it).
GitHub says that in the Artificial Analysis Coding Agent Index, Copilot Fusion Max scored 60 points, two points behind Claude Code with Opus 5.5 Max, at roughly half the average API cost per task, half the tokens, and around 40% less time per task. It is available on all Copilot plans within Visual Studio Code, the GitHub Copilot app, the Copilot CLI, and JetBrains IDEs, and for a limited time paid plans get a 30% discount on model costs when using it.
Source: GitHub (https://github.blog/changelog/2026-10-28-route-and-orchestrate-models-with-github-copilot-fusion/)
