Long-Horizon Terminal-Bench — leaderboard

Metric: Mean Score (%). Source: zli12321.github.io. 20 models tracked.

Top models

#ModelScore
1Grok 4.550.5
2Claude Sonnet 549.7
3Claude Opus 4.849.2
4Claude Fable 548.7
5GPT-5.6 Sol45.1
6GPT-5.544.5
7MiniMax-M338.5
8Claude Sonnet 4.637.3
9Kimi K2.7 Code36.7
10GLM-5.231.6
11Qwen 3.6 Plus31.3
12DeepSeek V4 Pro30.7
13Qwen 3.7 Max29.6
14Hy328.8
15Gemini 3.1 Pro (Preview)27.9

Interactive version: aibenchmarks.dev/benchmark?slug=long-horizon-terminal-bench · How the rankings work · Data refreshed daily, snapshot 2026-07-16.