| Rank | Model | Score |
|---|---|---|
| 1 | deepseek-v3-2-exp | 97.1 |
| 2 | deepseek-v3-1 | 93.4 |
| 3 | deepseek-r1-0528 | 92.3 |
| 4 | gemini-3-1-pro | 77.3 |
| 5 | gemini-3-pro | 72.1 |
| 6 | gpt-5-6-sol | 71.6 |
| 7 | gemini-3-flash | 68.7 |
| 8 | claude-fable-5 | 68.3 |
| 9 | gpt-4-5 | 62.5 |
| 10 | deepseek-v4-pro | 57 |
| 11 | gpt-5 | 54.9 |
| 12 | gemini-2-5-pro | 50.8 |
| 13 | o3 | 49.7 |
| 14 | o1 | 47 |
| 15 | gpt-5-1-instant | 45.6 |
| 16 | gpt-5-1-thinking | 45.6 |
| 17 | gpt-5-1 | 45.6 |
| 18 | inkling | 43.9 |
| 19 | o1-preview | 42.4 |
| 20 | kimi-k2-6 | 38.7 |
| 21 | gpt-4o | 38.2 |
| 22 | glm-5-2 | 38.1 |
| 23 | kimi-k2-5 | 36.9 |
| 24 | kimi-k2-base | 35.3 |
| 25 | nemotron-3-ultra-550b-a55b | 32.4 |
1 phaseActive