GitHub introduced HydraFusion Max on October 14, 2026, and updated how Copilot’s HydraFusion coordinates multiple models into one response. GitHub says Max shares HydraFusion’s architecture and workflow techniques but applies higher reasoning effort to demanding tasks while considering token efficiency. Sources: GitHub
HydraFusion now uses medium reasoning effort for drafts and high effort for reviews, replacing reliance on model defaults. Its model pool includes Claude Opus 5.5, GPT-6 Astra, GPT-6.1 Sol and GPT-6 Luna, with selection varying by workflow. Sources: GitHub
The Cascade workflow can add a judging step and request repairs when needed, according to GitHub. The Critique workflow uses tools to inspect the workspace more directly and runs critic models in parallel. Sources: GitHub
Both options remain research previews in VS Code, the Copilot app and Copilot CLI. They are available on Pro, Pro+, Business and Enterprise plans, with access governed by preview settings and organizational policies. GitHub’s announcement includes no numerical Max benchmarks, latency measurements, billing details, minimum client versions or general-availability date. Sources: GitHub