Devin Fusion Debuts: Multi-Model Coding Agent Scores 62, First on AA Index
AccBalanced · x · 2026-09-12
Cognition released Devin Fusion, billed as the most efficient frontier harness, pairing a frontier lead model with a cost-efficient sidekick. Artificial Analysis independently benchmarked it — the first multi-model coding agent on their Coding Agent Index: Claude Fable 5.1 (xhigh) + SWE-2 (medium) scores 62 on v1.5, GPT-6 Astra (xhigh) + SWE-2 (medium) scores 59. Configurable base/sidekick models, speed, and reasoning level, it reportedly retains frontier performance while cutting costs.
More from coding & agent
- HABIT-Bench: Best agent memory system scores just 33.61% at inferring user habits — SJTU · 2026-10-28
- Memory Tax study: dedicated agent memory systems often lose to naive RAG — UIUC-CS · 2026-10-28
- Regehr's 'Piano Test': would you trust formally verified code with your life? — moyix · 2026-10-28
- Vibe-coded 'AI 浮生记' game satirizes AI startup life with compute black-market trading — vista8 · 2026-10-28
- Cursor ships Annotate in design mode: agents take visual prompts via on-page markup — soleio · 2026-10-28
- The legendary fake Go generics: Unicode Canadian Syllabics as angle brackets — rickasaurus · 2026-10-28