List your company

Terminal-based task execution on Terminal Bench 2.1

67.2ScoreKimi-K2.6
48.4853.3458.263.06Jun 12, 2026

Evaluation results

MethodLinks
Kimi-K2.6Parameter Count=1T-A32B2026.0667.2
GLM-5.1Parameter Count=744B-A40B2026.0659.3
Nemotron 3 UltraParameter Count=550B-A55B2026.0656.4
MiniMax-2.7Parameter Count=230B-A10B2026.0655.5
DS-v4-FlashParameter Count=284B-A13B2026.0654.2
Qwen-3.5Parameter Count=397B-17B2026.0649.9
DS-v4-ProParameter Count=1.6T-A49B2026.0649.2