Mathematical Reasoning on OlympiadBench (Pass@1 Accuracy, Avg Token Length)
62.7Pass@1 Accuracy
ExpThink
Updated 23d ago
Evaluation Results
| Method | Links | ||
|---|---|---|---|
2026.05 | 62.7 | 2,430 | |
2026.05 | 57.3 | 9,649.2 | |
2026.05 | 56.1 | 3,594.3 | |
2026.05 | 55.7 | 4,968.7 | |
2026.05 | 55.7 | 5,603 | |
2026.05 | 54.7 | 2,977.2 | |
2026.05 | 54.4 | 7,364.1 | |
2026.05 | 54.2 | 4,206.2 | |
2026.05 | 54.1 | 6,230.3 | |
2026.05 | 45.3 | 5,121.3 | |
2026.05 | 45.2 | 6,490 | |
2026.05 | 44.6 | 3,978.2 | |
2026.05 | 43 | 3,057.1 | |
2026.05 | 42.4 | 5,158.8 | |
2026.05 | 41.5 | 8,915.1 | |
2026.05 | 39.6 | 4,651.1 |