| Rank | Model | Score |
|---|---|---|
| 1 | mimo-v2-flash | 0.862 |
| 2 | o3 | 0.859 |
| 3 | qwen3-next-80b-a3b-instruct | 0.827 |
| 4 | qwen3-235b-a22b-thinking-2507 | 0.797 |
| 5 | qwen3-235b-a22b-instruct-2507 | 0.792 |
| 6 | gemini-2-5-pro | 0.79 |
| 7 | qwen3-vl-235b-a22b-instruct | 0.774 |
| 8 | o4-mini | 0.746 |
| 9 | nemotron-3-super-120b-a12b | 0.739 |
| 10 | sarvam-105b | 0.71 |
| 11 | gemini-2-5-flash | 0.686 |
| 12 | nemotron-3-nano-30b-a3b | 0.677 |
| 13 | qwen3-vl-32b-instruct | 0.647 |
| 14 | qwen3-next-80b-a3b-thinking | 0.623 |
| 15 | qwen3-vl-32b-thinking | 0.605 |
| 16 | qwen3-vl-30b-a3b-instruct | 0.585 |
| 17 | qwen3-235b-a22b | 0.584 |
| 18 | deepseek-r1-0528 | 0.58 |
| 19 | qwen3-vl-30b-a3b-thinking | 0.567 |
| 20 | qwen3-vl-8b-thinking | 0.511 |
| 21 | sarvam-30b | 0.49 |
| 22 | qwen3-vl-4b-thinking | 0.368 |
1 phaseActive