Anthropic · model

Claude Sonnet 5.5 (high with fallback) benchmark scores

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

As of 2026-10-07, Claude Sonnet 5.5 (high with fallback) (Anthropic) has measured scores on 1 benchmark and estimated scores on 2 more.

Benchmark	Score	Source
Terminal-Bench 2.1	89.1%	estimated ± 7.0 pp, medium confidence
Terminal-Bench 4.0	50.0%	estimated ± 8.2 pp, medium confidence
Terminal-Bench Science 0.1	34.3%	measured