Long-context Evaluation on RULER ultra-long context official (Context Length Sweep)
96Accuracy (128K)
Qwen3-Next-80B-A3B-Instruct
Updated 4d ago
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
2026.02 | 96 | 86.9 | 80.3 | - | |
| 93.9 | 90.9 | 84.5 | - | ||
2026.02 | 89.4 | 87.1 | 86.3 | 81.6 | |
2026.02 | 89.1 | 78.4 | 72.8 | - |