| # | Model | Provider | Type | Score | |
|---|---|---|---|---|---|
| 1 | Kimi K2.6 | Moonshot AI | Open | 96.9% | |
| 2 | Qwen3.6 Plus | Alibaba | Closed | 96.9% | |
| 3 | Qwen3.5 397B | Alibaba | Open | 95.8% | |
| 4 | Step 3.7 Flash | StepFun | Open | 95.3% | |
| 5 | Qwen3.6-27B | Alibaba | Open | 94.7% | |
| 6 | Qwen3.5-27B | Alibaba | Open | 93.7% | |
| 7 | Qwen3.5-122B-A10B | Alibaba | Open | 93.2% | |
| 8 | Qwen3.5-35B-A3B | Alibaba | Open | 92.7% | |
| 9 | Gemini 3 Pro | Closed | 88.0% | ||
| 10 | GPT-5.2 | OpenAI | Closed | 75.9% | |
| 11 | Claude Opus 4.5 | Anthropic | Closed | 67.0% |