hmmt Feb2025
7 models evaluated
|
| 1 | GLM-5 | Z.AI | Open | 97.5% | |
| 2 | Qwen3.6 Plus | Alibaba | Closed | 96.7% | |
| 3 | Kimi K2.5 | Moonshot AI | Open | 95.4% | |
| 4 | Qwen3.5 397B | Alibaba | Open | 94.8% | |
| 5 | Qwen3.6-27B | Alibaba | Open | 93.8% | |
| 6 | Claude Opus 4.5 | Anthropic | Closed | 92.9% | |
| 7 | Qwen3.6-35B-A3B | Alibaba | Open | 90.7% | |