Anthropic · model

Claude Opus 4.7 (max) benchmark scores

Name: benchgap: LLM benchmark scores, measured and estimated
Creator: benchgap

As of 2026-10-07, Claude Opus 4.7 (max) (Anthropic) has measured scores on 1 benchmark and estimated scores on 2 more.

Benchmark	Score	Source
τ³-Bench Banking	46.8%	estimated ± 2.6 pp, medium confidence
Harvey LAB	93.5%	estimated ± 1.6 pp, low confidence
ITBench SRE	46.7%	measured