Grok 4.6 ties GPT-5.6 on intelligence index with top-tier agentic performance at lower cost
ArtificialAnlys · x · 2026-08-12
Artificial Analysis released a detailed evaluation of Grok 4.6. The model scores 61 on the Intelligence Index, joining the frontier alongside GPT-5.6 Sol, trailing only Anthropic's Claude Opus 5 and Claude Fable 5.
Strong Agentic Performance
- Excels in the GDPval-AA v2 evaluation, second only to Claude Opus 5.
- Ranks among the top models in 𝜏³-Banking and Terminal-Bench v2.1.
- Demonstrates high turn-efficiency in the AA-Briefcase benchmark, completing tasks in 53 turns vs. 103 turns for Claude Opus 5.
Highly Cost-Effective
- Headline pricing remains unchanged from Grok 4.5 at $2/$6 per 1M input/output tokens.
- Priced 60%+ below Claude Opus 5 and GPT-5.6 Sol, placing it firmly on the Intelligence vs. Cost Pareto frontier.
Related event: Grok 4.6 Tops Benchmarks, Keeps Low Pricing(11 posts)→
More from Models
- Vercel AI Gateway Adds Grok and DeepSeek Models with Zero Markup — brandon_galang · 2026-08-13
- DeepSeek V4 Pro Now Available on OpenCode Go with Leaked Benchmarks — op7418 · 2026-08-13
- Grok 4.6 Tops Artificial Analysis Agentic Index, Tying Claude Opus 5 Max — XFreeze · 2026-08-13
- DeepSeek 4 Flash Price Wars Thrive as Community Awaits 4 Pro Pricing — AccBalanced · 2026-08-13
- Grok Build Rolls Out Version 4.6 to SuperGrok Heavy Subscribers — EricBuess · 2026-08-13
- Should You Use Strong Models for Simple Tasks? Fable Crushes Opus in Email Drafting — mattbeane · 2026-08-13