Grok 4.6 Tops Knowledge-Work, Productivity, and Legal Benchmarks

elonmusk · x · 2026-08-13

Elon Musk retweeted news regarding Grok 4.6's latest benchmark performance. The model reportedly leads on the two strongest knowledge-work and real-world productivity benchmarks (GDPVal-AA and AA-Briefcase), as well as on legal benchmarks, while remaining highly competitive in coding-agent and general intelligence metrics.

Original post →

More from Models

Models channel →