Long-context evaluation on RULER 4k context length (test)
25.36MK
GDN
Updated 16d ago
Evaluation Results
| Method | Links | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
2026.07 | 25.36 | 22.62 | 25.38 | 80.49 | 81.67 | 38.6 | 68.37 | 37.68 | 47.52 | |
2026.07 | 21.5 | 52.2 | 51.3 | 36.7 | 76.3 | 32.2 | 49.6 | 63.7 | 47.9 | |
2026.07 | 19.9 | 49.1 | 35 | 32.9 | 76.6 | 31.8 | 35.5 | 76.5 | 44.7 | |
2026.07 | 19.5 | 67.6 | 65 | 45.9 | 51.3 | 24.2 | 53.9 | 85.2 | 45.2 | |
2026.07 | 4.9 | 3.3 | 3.6 | 3 | 13.6 | 14.2 | 14 | 0.7 | 7.2 | |
2026.07 | 1.33 | 2.4 | 2.38 | 34.44 | 60.04 | 21.27 | 23.49 | 2.65 | 18.5 | |
2026.07 | 0 | 0 | 0 | 0.2 | 0.7 | 6.4 | 9.1 | 0 | 2.1 |