What is GitHub Copilot Fusion?
Copilot Fusion has been generally available since October 28, 2026. According to GitHub, selecting Copilot Fusion in the model picker does not choose a specific model but a system that routes and orchestrates models for each task. Developers no longer need to switch models during a task or decide which model suits each step. They only set the effort level.
According to GitHub, Fusion combines the multi-model orchestration of Project HydraFusion with Copilot's fine-tuned auto model selection. The goal is a balance between performance and cost efficiency. For each prompt, Fusion determines which models to use and how they work together, while developers keep control over the level of effort. Anyone who had enabled auto, HydraFusion or HydraFusion Max in the research preview is moved automatically to the corresponding Fusion effort level.
Which effort levels and workflows are there?
GitHub names three effort levels. It also describes how Fusion combines models when needed: on their own, as a cascade or with mutual review. Fusion decides which workflow to use for each task, and developers can watch how it analyzes the task and selects a model.
- Fusion Standard prioritizes versatile models for everyday coding.
- Fusion High prioritizes capable models for challenging tasks and deeper reasoning.
- Fusion Max prioritizes the most capable models for the most demanding tasks.
- Single: one model solves the task. Cascade: an efficient model drafts, and a quality check decides whether to accept it or escalate to a stronger model.
- Critique: one model drafts, a second reviews it, and the first revises.
What does GitHub say about performance and cost?
According to GitHub, Copilot Fusion Max scored 60 in the Artificial Analysis Coding Agent Index, within two points of Claude Code with Opus 5.5 Max. GitHub says it delivered that result at about half the average API cost per task, using about half as many tokens and taking about 40 percent less time per task.
Each model Fusion uses is billed at its standard rate. For a limited time, users on paid Copilot plans get a 30 percent discount on model costs when using Fusion. Our assessment, not a vendor statement: vendor benchmarks do not replace your own comparison with typical tasks from your code.
What can administrators control?
Copilot Fusion only uses models enabled in enterprise and organization model policies. Enterprise admins can see which models Fusion picks from, set each model's effective policy state independently and disable models for Fusion without disabling them for everyone. Using Enterprise managed settings, Fusion can be set as the default model at any effort level.
Our assessment, not a vendor statement: The option to set Fusion as the default makes clear rules important. Teams should define which level applies to everyday tasks and when Max is allowed, so costs stay predictable. A monthly look at which models Fusion actually used also helps, especially once the limited-time discount ends.
Where is Copilot Fusion available?
According to GitHub, Copilot Fusion is available on all Copilot plans in Visual Studio Code, the GitHub Copilot app, GitHub Copilot CLI and JetBrains IDEs. Support for more clients is coming. To start, update the client, select Copilot Fusion in the model picker and choose an effort level. As a practical example, GitHub quotes AT&T: there, Fusion Standard improved token efficiency because developers previously often chose a more powerful model for a complex task and then forgot to switch back for everyday work. According to AT&T, this reduced unnecessary spend and preserved capacity for higher-value work, and the routing decisions improved over time. Our assessment, not a vendor statement: many teams know exactly this pattern, so it is worth reviewing your own model usage with it in mind.
What should businesses do now?
Check which models are enabled in your Copilot policies, because Fusion only picks from those. Test Fusion Standard with one team for everyday tasks and compare costs and results with your current model choice. Then decide whether Fusion should become the default and who may use the Max level.
The free AI System Check helps organize AI tools, costs and approvals across your development team. These notes are suggestions, not customer results.
Keep it verifiable






