GitHub Copilot updates HydraFusion and adds HydraFusion Max, with Claude Opus 5.5 and the GPT-6 family in its model pool

GitHub updated HydraFusion, the Copilot research-preview feature that orchestrates multiple models into a single response, and introduced HydraFusion Max. The model pool now includes Claude Opus 5.5, GPT-6 Astra, GPT-6.1 Sol and GPT-6 Luna, with selection depending on the workflow. Reasoning effort is now tailored per stage (medium for initial drafts, high for review) instead of each model's default. The Cascade workflow adds a judging step that requests a repair only when the initial work needs it, and Critique now inspects the actual workspace with tools and uses parallel critic models rather than relying on text summaries.

HydraFusion Max uses the same architecture but applies higher effort where depth matters, targeting the most demanding tasks while orchestration still considers token efficiency. The changelog publishes no new benchmark or pricing figures. Both options work in VS Code, the GitHub Copilot app and Copilot CLI for Copilot Pro, Pro+, Business and Enterprise users with experimental or preview features enabled, subject to organization model policies.