GRPO
| Method | Links | |||
|---|---|---|---|---|
2026.05 | 78 | 59.2 | 71.16 | |
2026.05 | 77.49 | 62.1 | 71.89 | |
2026.05 | 77.26 | 59.2 | 70.69 | |
2026.05 | 76.67 | 60.2 | 70.68 | |
2026.05 | 76.34 | 58.9 | 70 | |
2026.05 | 76.23 | 58.9 | 69.93 | |
2026.05 | 75.94 | 59.3 | 69.89 | |
2026.05 | 52.51 | 13.3 | 38.25 | |
2026.05 | 43.43 | 12.9 | 32.33 | |
2026.05 | 37.33 | 14.64 | 27.02 | |
2026.05 | 36.8 | 12.24 | 25.64 | |
2026.05 | 35.73 | 11.84 | 24.87 | |
2026.05 | 35.67 | 12.32 | 25.05 | |
2026.05 | 34.2 | 10.88 | 23.6 | |
2026.05 | 26 | 12.8 | 21.2 | |
2026.05 | 25.31 | 12.2 | 20.55 | |
2026.05 | 25.09 | 11.2 | 20.04 | |
2026.05 | 24.91 | 14.8 | 21.24 | |
2026.05 | 24.51 | 9 | 18.87 | |
2026.05 | 23.89 | 10.7 | 19.09 | |
2026.05 | 23.66 | 8.7 | 18.22 |