Agentic WeeklyGet the weekly

GitHub Copilot Fusion is generally available and replaces Auto with Standard, High and Max effort levels

Release Ruler for this brief: marked at its publication date, Oct 28, 2026, among the AI models we track.

On October 28, 2026 GitHub made Copilot Fusion generally available: instead of picking a model, users pick Fusion Standard, High or Max and Copilot decides which models to use per task, billed at each model's standard rate with a 30% discount for a limited time on paid plans.

By Agentic Weekly · Evidence: Reported · How we work

First seen
(via GitHub changelog: Copilot)
Source published
(we saw it 4 minutes later)
Brief published

What changed

  • GitHub's October 28, 2026 changelog says Copilot Fusion is generally available on all Copilot plans in VS Code, the GitHub Copilot app, Copilot CLI and JetBrains IDEs. It combines Project HydraFusion's multi-model orchestration with Copilot's auto model selection.
  • Users choose an effort level rather than a model: Standard for everyday coding, High for challenging tasks and Max for the most demanding ones. Fusion then runs one model directly, a Cascade (a cheap draft escalated if a quality check fails) or a Critique loop.
  • GitHub reports that 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 and about 40% less time per task.
  • Billing is per model used at each model's standard rate, with a 30% discount on model costs for a limited time on paid plans. The JetBrains changelog says Fusion replaces the current Auto model option and is rolling out gradually.
  • Fusion only uses models enabled by enterprise and organization policies. Admins can exclude a model from Fusion without disabling it for everyone, and can set Fusion at any effort level as the default model.
  • A separate experimental preview lets Fusion Standard route between cloud models and on-device models such as MAI Code 1.1 Flash in Copilot CLI, the Copilot app and VS Code; routing decisions are still made in the cloud.

Primary source: GitHub changelog: Route and orchestrate models with GitHub Copilot Fusion. We are the fast, independent index; the vendor’s own page is the source of record for this release.

In the source’s words

“For each prompt, Copilot Fusion determines which models to use and how they work together, while you choose the right level of effort.”

“Administrators control which models Copilot Fusion can use and whether Copilot Fusion is the default.”

“Copilot Fusion still makes routing decisions in the cloud, and local models may have different safety protections from Copilot-provided cloud models.”

Early verdict

Copilot users who switch models by hand can try Fusion Standard for everyday work; enterprise admins should review which models are enabled before making Fusion the default, since it bills per model used.

Basis: Based only on the effort levels, admin controls and billing terms in GitHub's changelogs; we have not run Fusion.

What we have not verified

  • We did not run Copilot Fusion at any effort level.
  • The Artificial Analysis Coding Agent Index score, cost and time figures are GitHub's own reported numbers; we did not check them against Artificial Analysis.
  • The changelog does not say how long the 30% discount lasts beyond 'a limited time' (the JetBrains post says the first 30 days) or which models each effort level can pick.

This brief is Reported: written from the sources below, not from running the release ourselves.

Sources

Spotted an error? See our corrections policy or email the address on the contact page. Markdown version for agents: /news/github-copilot-fusion-generally-available.md

Want these briefs in your Google Top Stories? Add Agentic Weekly as a preferred source on Google.