GitHub launches HydraFusion, a Copilot preview routing one task across multiple models at 36-67% lower cost
shashib · x · 2026-09-18
- GitHub shipped HydraFusion on Sept 4, 2026 as a research preview inside Copilot — routing software, not a new frontier model. Developers pick HydraFusion instead of a single model, and Copilot's runtime decides how many models to run and in what order.
- Three paths: a mid-tier model finishes the task, escalate to a stronger one, or have a second model review the first draft.
- Estimated cost runs 36%-67% lower than Opus 5 on GitHub's three benchmarks.
- Motivation: with many models in Copilot, everyone picking the strongest blows up the token bill of flat plans. Auto selection still sent the whole task to one model; HydraFusion moves mid-task model switching into the runtime. The author argues enterprises will increasingly buy "when to spend more inference" plans rather than named models.
More from coding & agent
- Quota reset night: Codex back to 100% while Claude shows just 10% left — rudrank · 2026-11-04
- Self-improving software is inevitable: closing the agent-observability loop — bibryam · 2026-11-04
- Building a 7-GPU 132GB VRAM local inference rig: consolidate on EPYC or split across two systems? — infectiousstupidity · 2026-11-04
- Founder of 50-person, 12-country team says he talks to AI agents more than humans — Scobleizer · 2026-11-04
- Pi launches a minimal agent harness; Zammoth adds AI engineer to Mac music app — solyarisoftware · 2026-11-04
- Reply to Karpathy goes viral again: devs who refuse AI coding will be replaced first — CtrlAltDwayne · 2026-11-04