Grok 4.7's AA-Briefcase analytical quality Elo jumps to 1994 from 1690

ArtificialAnlys · x · 2026-09-22

Artificial Analysis breaks down Grok 4.7's agentic knowledge work scores: 1657 Elo on AA-Briefcase (+111 over Grok 4.6 high), just behind Claude Opus 5 and Claude Fable 5.1. Gains are driven by analytical quality: 1994 Elo vs 1690, while presentation quality slipped slightly to 1499 from 1519. The benchmark also checks deliverables against task requirements. On GDPval-AA (documents, spreadsheets, slides), it scores 1695 vs 1605 for the predecessor.

Related event: Grok 4.7 Scores 46 on Intelligence Index, Enters Top Four Labs at Double Token Cost(7 posts)→

Original post →

More from Models

Models channel →