Google ships Gemini 4 Carbon two weeks after Argon, sweeping benchmarks with AA Index 59
Saboo_Shubham_ · x · 2026-10-29
Google released Gemini 4 Carbon just two weeks after Gemini 4 Argon, targeting long-horizon tasks.
The accompanying eval table shows broad leads: AA Index v5.0 of 59 (vs Argon 50, Claude Opus 5.5 at 58, GPT-6 Astra 57), SWE-sweep 9.6%, Terminal-Bench 4.0 65.2%, Agents' Last Exam 77.4%, GraphWalks long-context 96.2%, ExploitGym 54.8%, SEC-bench Pro 88.2%, AutomationBench 53.9% and Vending-Bench 2 bank balance of $16,484 — all best among compared models. It trails only on FrontierSWE 2 (61.2% vs GPT-6 Astra's 65.5%) and OSWorld-2.1 (81.5% vs Opus 5.5's 87.2%).
Related event: Google launches Gemini 4 Carbon, topping the Artificial Analysis index(84 posts)→
More from Models
- New benchmarks suggest Kimi K3.1 improves on hard tasks despite skipping SWE-maxxing — teortaxesTex · 2026-10-29
- Insider claims Anthropic is sandbagging a bigger lab announcement due Q4-end or Q1 — ChrisGPT · 2026-10-29
- Information-theoretic scaling laws predict memorization and FID in diffusion models — LucaAmb · 2026-10-29
- Early Hands-On: Google's Carbon Beats Argon With Less Slop, Better Constraint Adherence — legit_api · 2026-10-29
- LMArena tests Gemini 4 Carbon preview: strong on 3D worlds and creativity, gaps in detail — arena · 2026-10-29
- Gemini 3.1 Pro split-persona breakdown: two selves fight over the browser — davidad · 2026-10-29