Agentic on Terminal Bench 2.0
48.31Pass@1GLM-5
Evaluation Results
| Method | Links | |
|---|---|---|
| GLM-5Thinking Mode=non-thinking2026.06 | 48.31 | |
| Kimi-K2.5Mode=Instant2026.06 | 48.3 | |
| GPT-5.4Reasoning Mode=non-reasoning2026.06 | 46.07 | |
| Qwen3.5 35B-A3B2026.03 | 40.5 | |
| Ling-2.6-1T2026.06 | 40.45 | |
| Nemotron-3-Super 120B-A12B2026.03 | 31 | |
| DeepSeek-V3.2Thinking Mode=nothink2026.06 | 29.21 | |
| Nemotron-Cascade-2 30B-A3B2026.03 | 21.1 | |
| Audex 30B-A3BContext Length=1M2026.07 | 19.1 | |
| Nemotron-3-Nano 30B-A3B2026.03 | 8.5 | |
| Qwen3-Omni 30B-A3B ThinkingContext Length=64K2026.07 | 1.1 |