SelfBench - terminal-bench/terminal-bench-2-1: leaderboard

Metric: Task success (%) on 87 tasks of the terminal-bench/terminal-bench-2-1 benchmark, run by SelfBench. Source: selfbench.dev. Saturation forecast: Around February 2027. 2 models tracked. Updated 18 Nov 2026.

Top models

#ModelSettingScoreOverall rank
1Dari RouterDefault83.91
2GPT-6 LunaLow65.52#102

No result here: #1 Claude Fable 5.2, #2 GPT-6.2 Astra, #3 Gemini 4 Carbon.

Interactive version: theaggregate.ai/benchmark?slug=selfbench-terminal-bench-terminal-bench-2-1 · How It Works · Data refreshed daily, snapshot 2026-11-18.