Reasoning and Code Generation Suite (MATH, GSM8K, MBPP, TheoremQA, BBH) (test)
54.38MATH Accuracy
FlexSwitch
Updated 1mo ago
Evaluation Results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
2026.04 | 54.38 | 731.57 | 84.84 | 60.2 | 24.5 | 73.47 | 59.48 | |
2026.04 | 53.76 | - | 84.31 | 59.2 | 25.75 | 73.69 | 59.34 | |
2026.04 | 48.22 | - | 78.32 | 51 | 26.12 | 55.85 | 51.9 | |
2026.04 | 47.8 | 5.21 | 79.91 | 51.4 | 26.38 | 55.73 | 52.24 | |
2026.04 | 47 | 35,254.31 | 82.41 | 60 | 12.5 | 71.55 | 54.69 | |
2026.04 | 43.9 | 12,255.67 | 76.95 | 49.8 | 24.25 | 55.38 | 50.06 |