| Rank | Model | Score |
|---|---|---|
| 1 | gpt-5-5 | 66.3 |
| 2 | claude-sonnet-4-6 | 64.4 |
| 3 | claude-opus-4-6 | 63.5 |
| 4 | claude-sonnet-4-6 | 63.2 |
| 5 | gpt-5-5 | 62.9 |
| 6 | claude-sonnet-4-6 | 62.3 |
| 7 | gpt-5-5 | 62 |
| 8 | gpt-5-5 | 61.4 |
| 9 | claude-opus-4-6 | 60.2 |
| 10 | deepseek-v4-pro-max | 58.8 |
| 11 | deepseek-v4-pro-max | 58.6 |
| 12 | claude-opus-4-6 | 57.3 |
| 13 | deepseek-v4-pro-max | 56.2 |
| 14 | kimi-k2-6 | 56.1 |
| 15 | minimax-m3 | 52.5 |
| 16 | minimax-m3 | 52 |
| 17 | kimi-k2-6 | 51.9 |
| 18 | kimi-k2-6 | 51.5 |
| 19 | claude-haiku-4-5 | 51.3 |
| 20 | claude-haiku-4-5 | 50.7 |
| 21 | claude-sonnet-4-6 | 45.8 |
| 22 | claude-haiku-4-5 | 42.9 |
| 23 | claude-opus-4-6 | 42.5 |
| 24 | minimax-m3 | 42.1 |
| 25 | claude-haiku-4-5 | 40.5 |
1 phaseActive