Mathematical Reasoning on AIME 2025 (Accuracy and Token Costs)
40Sample Count
SC
Updated 3d ago
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
2026.02 | 40 | - | 35.6 | 6.67 | |
2026.02 | 40 | - | 37.3 | 16.67 | |
2026.02 | 40 | - | 164.1 | 16.67 | |
2026.02 | 39.83 | - | 37.8 | 6.67 | |
2026.02 | 37.9 | 4.9 | 35.1 | 6.67 | |
2026.02 | 37.64 | - | 37.4 | 16.67 | |
2026.02 | 36.73 | - | 42.4 | 6.67 | |
2026.02 | 33.86 | 5.2 | 32.7 | 16.67 | |
2026.02 | 32.72 | - | 136.3 | 16.67 | |
2026.02 | 32.58 | - | 39.3 | 16.67 | |
2026.02 | 30.13 | - | 29.4 | 10 | |
2026.02 | 28.96 | 17.7 | 119.8 | 16.67 | |
2026.02 | 27.12 | - | 119.1 | 16.67 | |
2026.02 | 26.2 | - | 27.6 | 13.33 | |
2026.02 | 17.9 | - | 74.5 | 16.67 |