List your company

End-to-end terminal tasks on Terminal-Bench 2

49.6ScoreGPT-5
-1.98411.40824.838.192Dec 4, 2025

Evaluation results

MethodLinks
GPT-5Model Type=Proprietary2025.1249.6
Claude-Sonnet-4.5Model Type=Proprietary2025.1242.8
Kimi-K2-thinkingModel Type=Open Source, Parameter Scale=> 100B2025.1235.7
Gemini-2.5-proModel Type=Proprietary2025.1232.6
DeepSeek-V3.1-Nex-N1Model Type=Open Source, Parameter Scale=> 100B2025.1231.8
Minimax-M2Model Type=Open Source, Parameter Scale=> 100B2025.1230
GLM-4.6Model Type=Open Source, Parameter Scale=> 100B2025.1224.5
DeepSeek-V3.1Model Type=Open Source, Parameter Scale=> 100B2025.1222.2
Qwen3-32B-Nex-N1Model Type=Open Source, Parameter Scale=< 100B2025.1216.7
Qwen3-30B-A3B-Nex-N1Model Type=Open Source, Parameter Scale=< 100B2025.128.3
Qwen3-32BModel Type=Open Source, Parameter Scale=< 100B2025.127.9
Qwen3-30B-A3BModel Type=Open Source, Parameter Scale=< 100B2025.126
InternLM3-8B-Nex-N1Model Type=Open Source, Parameter Scale=< 100B2025.120