Long-context language tasks (MC, QA, Sum) on ∞Bench
78.6MC Accuracy
RAG
Updated 4d ago
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
2026.01 | 78.6 | 29.3 | 25.1 | |
2026.01 | 72.5 | 30.7 | 31.7 | |
2026.01 | 72 | 30 | 31.2 | |
2026.01 | 71.6 | 34.7 | 29.7 | |
2026.01 | 70.7 | 31.3 | 30.2 | |
2026.01 | 70.7 | 30.8 | 30.5 | |
2026.01 | 70.3 | 38.2 | 31 | |
2026.01 | 65.5 | 33.6 | 28 | |
2026.01 | 64.2 | 29.3 | 26 | |
2026.01 | 63.8 | 28.2 | 28.8 | |
2026.01 | 63.3 | 27.1 | 27.7 | |
2026.01 | 62.9 | 27.1 | 28.1 | |
2026.01 | 62 | 27.6 | 27.3 |