| # | Model | Provider | Type | Score | |
|---|---|---|---|---|---|
| 1 | Gemini 3 Pro | Closed | 87.6% | ||
| 2 | Kimi K2.5 | Moonshot AI | Open | 86.6% | |
| 3 | Qwen3.5 397B | Alibaba | Open | 84.7% | |
| 4 | MiniMax M3 | MiniMax | Open | 84.6% | |
| 5 | Claude Opus 4.5 | Anthropic | Closed | 84.4% | |
| 6 | Qwen3.6-27B | Alibaba | Open | 84.4% | |
| 7 | Qwen3.6 Plus | Alibaba | Closed | 84.0% | |
| 8 | Qwen3.6-35B-A3B | Alibaba | Open | 83.7% |