| # | Model | Provider | Type | Score | |
|---|---|---|---|---|---|
| 1 | Qwen3.8 Max | Alibaba | Open weights | 92.1% | |
| 2 | MiniMax M3 | MiniMax | Open weights | 91.6% | |
| 3 | Qwen3.7 Plus | Alibaba | Proprietary | 91.4% | |
| 4 | Qwen3.8-27B | Alibaba · Open weight | 91.1% | ||
| 5 | Qwen3.6-35B-A3B | Alibaba · Open weight | 89.9% | ||
| 6 | Muse Glimmer 30B | Meta · Open weight | 75.8% |