List your company

Command-line Interface Tasks on Terminal-Bench 2.0

57.3Terminus2 JSON ScoreClaude-Opus-4.5
31.61238.28144.9551.619Feb 28, 2026

Evaluation results

MethodLinks
Claude-Opus-4.5Size=?2026.0257.358.453.951.7
Claude-Sonnet-4.5Size=?2026.0251.751.741.637.1
Kimi-K2.5Size=1000A322026.0249.438.8927.5
DeepSeek-V3.2Size=671A372026.0239.334.8——
GLM-4.7Size=358A322026.0237.144.9—31.5
Qwen3-Coder-NextSize=80A32026.0236.234.230.925.8
MiniMax-M2.1Size=230A102026.0232.6—42.739.3