Science Reasoning on GPQA Diamond (Accuracy, Avg., Drop)
56.06Accuracy
BF16 Baseline
Updated 3d ago
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
2026.01 | 56.06 | 70.63 | - | |
2026.01 | 48.48 | 60.78 | -9.85 | |
2026.01 | 47.47 | 58.28 | -12.35 | |
2026.01 | 36.87 | 48.72 | - | |
2026.01 | 32.83 | 41.31 | -7.41 | |
2026.01 | 31.82 | 38.39 | -10.33 | |
2026.01 | 29.8 | 17.34 | -23.76 | |
2026.01 | 28.45 | 41.1 | - | |
2026.01 | 26.94 | 21.44 | -19.66 | |
2026.01 | 24.24 | 4.84 | -36.26 | |
2026.01 | 8.59 | 2.11 | -46.61 |