Terminal-Bench 4.0 | Results

Benchmark: Terminal-Bench
Evaluated by: Terminal-Bench Team
Release date: August 28, 2026
Version: Terminal-Bench 4.0
Metric: Terminal-Bench Accuracy

Terminal-Bench 4.0 results
Provider	Model	Score	Source	Registry No.
Anthropic	Claude Sonnet 5.5 (max)	70.6%	Source for Claude Sonnet 5.5 (max) on Terminal-Bench 4.0 (opens in a new tab)	20016
Anthropic	Claude Opus 5.5 (xhigh)	66.4%	Source for Claude Opus 5.5 (xhigh) on Terminal-Bench 4.0 (opens in a new tab)	20015
Anthropic	Claude Fable 5.1	55.8%	Source for Claude Fable 5.1 on Terminal-Bench 4.0 (opens in a new tab)	20004