GitHub Copilot Fusion tops Terminal-Bench at 71.2%, costing 57-88% less than frontier models
mariorod1 · x · 2026-11-18
Announced at Microsoft Ignite, GitHub Copilot Fusion scored 71.2% on Terminal-Bench 4.1 — 5.4 points ahead of frontier models Opus 5.5 and Fable 5.2 — while costing 57-88% less. Fusion is a runtime model router/orchestrator: instead of picking a model, it auto-selects models, reasoning effort, and one of three execution patterns (Single, Cascade with judge escalation, parallel Critique), routing along cache boundaries and factoring real-time system health to cut cost.
More from coding & agent
- YC-backed Ample launches one-stop cloud for agents, hits 5,000 users two weeks after launch — ycombinator · 2026-11-18
- OmniLoRA: train any model on any GPU inside your ComfyUI graph — pente5 · 2026-11-18
- findatasets ships /feedback endpoint so AI agents can file bug reports, following Coinbase's lead — virattt · 2026-11-18
- Put rules in code, not prose: what actually makes an AI instruction a Rule — gerardsans · 2026-11-18
- Dev shares a tiered file-permission policy to stop agents stalling on approvals — gaganghotra_ · 2026-11-18
- Museo, a whisperable AI ring for phone-free walks, ships next month with open dev platform — Scobleizer · 2026-11-18