Cognition: Gemini 4 Neon scores 46.2% on FrontierCode 2.0, near Opus 5.5 at 1/3 cost
scaling01 · x · 2026-11-20
- Cognition evaluated Gemini 4 Neon on FrontierCode 2.0, its benchmark measuring whether you should merge the PR
- Scored 46.2%, within one point of Opus 5.5, and the strongest of the Gemini series
- Cost: $4.38 per task, about a third of Opus 5.5's
- Now available in Devin Desktop and Devin CLI
Widely read as evidence Google is back at the frontier for coding models.
More from coding & agent
- Sentry ran 10k+ AI evals in 90 days, over half in the past month alone — zeeg · 2026-11-20
- Free skill uses Blender and Fable 5.2 to un-slop vibe-coded websites from just a URL — nateliason · 2026-11-20
- Gemini API adds tool search preview: defer tool loading so agents skip sending full tool libraries — rseroter · 2026-11-20
- Builder on Reddit: Claude removed the friction between me and my tools — Live-Fault1571 · 2026-11-20
- HKU Open-Sources PhoneGo, a CLI+GUI Hybrid Phone Agent Cutting Steps by 47% and Latency by 56% — aigclink · 2026-11-20
- Anthropic's AI code review: real-issue detection jumped from 16% to 54% — dotey · 2026-11-20