Grok 4.7 (High): benchmark results

Provider: xAI. Access: API.

Unified ELO 1752 ± 1, rank #26 of 3040 rated models, from 61 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
AutomationBench-AA - Finance - Raw Objective Completion92.19Raw Objective Completion (%)99.4
AA-Briefcase - DOCX - Rubric Pass Rate57.99Rubric Pass Rate (%)98.8
AutomationBench-AA - Strict37.75Strict Task Success (%)98.5
AA GDPval1693.72ELO98.3
AA-Omniscience Index - Science, Engineering & Mathematics26.4Omniscience Index97.9
AA-Briefcase - Analytical Quality Elo1967.4Elo97.7
AutomationBench-AA - Support - Raw Objective Completion90.97Raw Objective Completion (%)97.6
AA-Omniscience Index - Business25.8Omniscience Index97.5
Artificial Analysis Intelligence Index46.33Intelligence Index97.5
AA-Omniscience Index - Humanities & Social Sciences31.1Omniscience Index97.2
AA Omniscience30.9Score96.8
AA-Omniscience Index - Health14.8Omniscience Index96.6

Interactive version: theaggregate.ai/model?slug=grok-4-7-high · How It Works · Data refreshed daily, snapshot 2026-09-22.