Mathematical Reasoning on GSM-Hard (Accuracy, AVG, Improvement Overhead)
89.52Accuracy
GPT-5-High*
Updated 26d ago
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
2025.09 | 89.52 | - | - | |
2025.08 | 72.1 | 78.22 | - | |
| 69.89 | 74.37 | 6.5 | ||
| 68.75 | 74.18 | 6.2 | ||
2025.08 | 68.08 | 68.52 | - | |
2026.03 | 64 | - | - | |
2025.08 | 63.99 | 69.86 | - | |
| 63.3 | - | - | ||
| 60.23 | 63.59 | 9.5 | ||
| 57.4 | - | - | ||
2025.08 | 56.48 | 58.09 | - | |
| 52.01 | 58.46 | 0.63 | ||
2025.09 | 48.1 | - | - | |
2026.03 | 44.6 | 66.01 | - | |
2026.03 | 42.5 | 64.99 | - | |
2026.03 | 41.25 | 61.15 | - | |
2026.03 | 39.8 | - | - | |
2026.03 | 39.6 | - | - | |
2026.03 | 39.6 | - | - | |
2026.03 | 39.6 | - | - | |
2026.03 | 39.4 | 57.01 | - | |
2026.03 | 39.4 | - | - | |
2026.03 | 39.4 | - | - | |
2026.03 | 38 | 60.33 | - | |
2025.08 | 36.69 | 55.48 | - | |
2026.03 | 35.8 | 65.94 | - | |
2026.03 | 34.8 | - | - | |
2026.03 | 34.6 | - | - | |
2026.03 | 34.2 | 68.23 | - | |
2026.03 | 33.8 | - | - | |
2026.03 | 33.2 | 59.56 | - | |
2026.03 | 33 | 61.13 | - | |
2026.03 | 32.8 | - | - | |
2026.03 | 31.8 | 64.19 | - | |
2026.03 | 31.8 | - | - | |
2026.03 | 31.8 | - | - | |
2026.03 | 31.4 | - | - | |
2026.03 | 31.2 | - | - | |
2026.03 | 31.2 | - | - | |
2026.03 | 31 | - | - | |
2026.03 | 30 | - | - | |
2026.03 | 29.8 | - | - | |
2026.03 | 23.4 | - | - | |
2026.03 | 23 | - | - | |
2026.03 | 22.8 | - | - | |
2026.03 | 21.4 | - | - |