Gemini 4 Carbon (High) from Google DeepMind debuted on Arena's leaderboards, ranking #1 in Text Arena with 1528 points, ahead of Gemini 4 Argon (High) at 1520. In Agent Arena it placed #2 with a +16.2% net improvement across +21.9K real-world agentic sessions, surpassing GPT-6 Astra and Fable-5.1, at $1.80 median cost per task—72% less than #1 Claude Opus 5.5 (Max). In Code Arena: WebDev it scored 1768 points for #4, two points behind Claude Sonnet 5.5 (xHigh), ranking #3 in Gaming and Simulations and #5 in Brand & Marketing. The author says it reshapes the Pareto frontier across all three arenas.

AI summary of the linked source — not editorial content.

Story signal · how it spread on X · via Elfa

as of

Related on X Google launches Gemini 4 Carbon with cybersecurity and long-horizon capability gains · 65 accounts · first post

Explore the topic

Loading related coverage…

Comments

Loading comments…