Terminal Capability Evaluation on Terminal-Bench 2.0
27.4AccuracyNemotron-T-32B
Evaluation Results
| Method | Links | |
|---|---|---|
| Nemotron-T-32BModel Category=Open Source Models, Size=32B2026.02 | 27.4 | |
| GPT-5-MiniModel Category=Closed Source Models2026.02 | 24 | |
| Qwen3-CoderModel Category=Open Source Models, Size=480B2026.02 | 23.9 | |
| Grok 4Model Category=Closed Source Models2026.02 | 23.1 | |
| Qwen3-Max-ThinkingModel Category=Closed Source Models2026.02 | 22.5 | |
| Nemotron-T-14BModel Category=Open Source Models, Size=14B2026.02 | 20.2 | |
| GPT-OSS (high) 120BModel Category=Open Source Models, Size=120B2026.02 | 18.7 | |
| Gemini 2.5 FlashModel Category=Closed Source Models2026.02 | 16.9 | |
| Grok Code Fast 1Model Category=Closed Source Models2026.02 | 14.2 | |
| GPT-5-NanoModel Category=Closed Source Models2026.02 | 7.9 | |
| Qwen3-32BModel Category=Open Source Models, Size=32B2026.02 | 3.37 | |
| GPT-OSS (high) 20BModel Category=Open Source Models, Size=20B2026.02 | 3.1 |