Gemini 4 Carbon (High) tops AA Intelligence Index at 59, but costs $7.76 per task
ArtificialAnlys · x · 2026-10-29
Artificial Analysis published its full analysis page for Gemini 4 Carbon (High):
- Intelligence Index score of 59, ranked #1 of 52, well above the class median of 42
- Expensive: $4.00/M input tokens (median $2.00) and $20.00/M output tokens (median $10.00); evaluation costs $7.76 per Intelligence Index task
- Extremely verbose: generated 200M output tokens vs a 64M median, ranking 49/51 on verbosity
- Specs: text/image/speech/video input, 1M-token context window, reasoning enabled
Bottom line: top-tier intelligence, but poor price-performance versus peers and verbosity that inflates real-world cost.
Related event: Hassabis hypes Gemini 4 Carbon as SOTA across key benchmarks(87 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