GitHub said Wednesday that GitHub Copilot Fusion is generally available, letting users choose Fusion instead of picking a model for each task so Copilot can route and orchestrate models^1,2.

The company’s changelog said Fusion combines Project HydraFusion’s multi-model orchestration with Copilot auto model selection². Techmeme, summarizing Meredith Shubel in The New Stack, also said GitHub released the GitHub Copilot app alongside the Fusion news¹.

In the changelog, GitHub said selecting Copilot Fusion in the model picker chooses a system that decides which models to use and how they work together, while the user sets an effort level: Standard for everyday coding, High for harder tasks, or Max for the most demanding work².

GitHub described three execution patterns: Single, in which one model handles the task; Cascade, in which an efficient model drafts and a check may escalate to a stronger model; and Critique, in which one model drafts, another reviews, and the original revises².

GitHub said that since HydraFusion launched, Copilot Fusion Max scored 60 on the Artificial Analysis Coding Agent Index, within two points of Claude Code with Opus 5.5 Max, at about half the average API cost per task, using about half as many tokens and about 40% less time per task². Those figures come from GitHub and were not independently verified in the sources reviewed.

Administrators can limit which models Fusion may use through enterprise and organization model policies and can set Fusion at any effort level as the default for users with enterprise managed settings, GitHub said². Fusion is billed at each model’s standard rate, with a limited-time 30% discount on model costs for paid Copilot plans².

GitHub quoted AT&T principal software engineer Hari Pokhrel as saying Fusion Standard “has improved token efficiency across AT&T by automatically selecting the right model”². The company said Fusion is on all Copilot plans in Visual Studio Code, the GitHub Copilot app, Copilot CLI, and JetBrains IDEs, with more clients later².