Long-context Language Modeling on Composite Suite (MRCR v2, GraphWalks, LongBench v2, RULER, AA-LCR)
78.7Average Score
IndexCache
Updated 1mo ago
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
2026.03 | 78.7 | 72.3 | 90.8 | 66 | 97.3 | 67.2 | |
2026.03 | 78.4 | 71.1 | 92.7 | 64.5 | 97.7 | 66.2 | |
2026.03 | 78.1 | 72.8 | 90.2 | 65.1 | 97.6 | 64.6 | |
2026.03 | 78 | 70.8 | 90.3 | 63.7 | 97.6 | 67.6 | |
2026.03 | 72.7 | 65.8 | 74.9 | 62.2 | 96.2 | 64.6 |