HydraFusion Adds Max Option and Upgrades Model Pool

What changed?

HydraFusion, GitHub's multi-model orchestration system for Copilot, now draws from a pool that includes frontier models: Claude Opus 5.5, GPT-6 Astra, GPT-6.1 Sol, and GPT-6 Luna. The orchestration logic has been updated to explicitly tailor the reasoning effort to each workflow stage, using medium effort for initial drafts and high effort for reviews. Cascade workflows can now judge initial outputs before escalating to more complex fixes, and Critique workflows leverage tools for workspace inspection and employ parallel critic models.

A new option, HydraFusion Max, is introduced for tasks requiring in-depth analysis. Max applies higher reasoning efforts while keeping token usage efficient via orchestration strategies. Both HydraFusion and HydraFusion Max are accessible in VS Code, the GitHub Copilot app, and Copilot CLI on Pro, Pro+, Business, and Enterprise plans with preview features enabled.

Why does it matter to an everyday developer?

Developers now benefit from the strengths of several state-of-the-art models without needing to select each one manually. Reasoning is more finely tuned for the actual task phase, improving draft efficiency and review depth. The Cascade and Critique workflow improvements mean more accurate code generation and review, especially for complex or ambiguous requests. For advanced use cases, HydraFusion Max enables deeper analysis, important for high-stakes, complex coding tasks.

What can the developer do now?

Developers on eligible Copilot plans can begin using HydraFusion and HydraFusion Max in their preferred Copilot client by enabling experimental or preview features, subject to organizational settings. Use HydraFusion for everyday agentic tasks and switch to HydraFusion Max when higher reasoning and thoroughness are essential. No special configuration is required to leverage the updated model pool or new workflow optimizations; they are active with the HydraFusion options.