Google sweeps top 3 on Arena text leaderboard as Gemini 4 Carbon scores 1527
demishassabis · x · 2026-11-23
Google DeepMind CEO Demis Hassabis retweeted the latest LMArena text leaderboard, where Google now holds the top three spots: Gemini 4 Carbon (1527), Gemini 4 Argon (1520), and Gemini 4 Neon (1513), ahead of Meta's Muse Wonder (1511) and Claude Opus 5.5 (1506). The screenshot also lists per-task costs: Gemini 4 Argon runs just $0.03/task while Claude Fable 5.1 (Max) costs $0.46, with top models now separated by only single-digit points.
More from Models
- Thinking Machines releases Inkling 2 Small, scoring 31 on AA Intelligence Index — _Hao_Zhu · 2026-11-23
- Leaked OpenAI loss ladder: unknown capabilities await at 0.68-0.7 — tallmetommy · 2026-11-23
- Developers heap praise on Gemini 4 Neon's 'most tasteful' writing — gajesh · 2026-11-23
- METR-fit extrapolation implies 66-hour task horizons for 6.2 Astra, 128 hours for internal model — eliebakouch · 2026-11-23
- $200 Google AI Ultra subscription quietly includes a 96GB Blackwell GPU for 10 days — kieranklaassen · 2026-11-23
- Tesla FSD insider: end-to-end is settled, winning hinges on data diet quality — Singularitarian · 2026-11-23