Benchmark
End-to-end tasks completed in a real terminal environment (v2.1 task set).
Score history →| # | Model | Score |
|---|---|---|
| 1 | DeepSeek V4.1-Flash 0910 | 90.6% |
| 2 | GPT-6 Astra | 89.9% |
| 3 | Claude Opus 5 | 89.1% |
| 4 | GPT 5.6-Sol | 88.8% |
| 5 | Kimi K3 | 88.3% |
| 6 | GLM 5.3 | 88.2% |
| 7 | DeepSeek V4-Pro 0813 | 87.9% |
| 8 | DeepSeek V4-Flash 0731 | 82.7% |
| 9 | MiniMax M3 | 66% |