| # | Model | Provider | Type | Score | |
|---|---|---|---|---|---|
| 1 | Qwen3.5 397B | Alibaba | Open | 88.6% | |
| 2 | Qwen3.6 Plus | Alibaba | Closed | 88.0% | |
| 3 | Kimi K2.6 | Moonshot AI | Open | 87.4% | |
| 4 | Gemini 3 Pro | Closed | 86.6% | ||
| 5 | Qwen3.5-122B-A10B | Alibaba | Open | 86.2% | |
| 6 | Qwen3.5-27B | Alibaba | Open | 86.0% | |
| 7 | Qwen3.5-35B-A3B | Alibaba | Open | 83.9% | |
| 8 | GPT-5.2 | OpenAI | Closed | 83.0% | |
| 9 | Claude Opus 4.5 | Anthropic | Closed | 74.3% |