Long-context language understanding on InfiniteBench (QA Subset)
7.84InfiniteBench QA (EN) ScoreQwen3-1.7B-ALLMEM
Evaluation results
| Method | Links | ||
|---|---|---|---|
| Qwen3-1.7B-ALLMEMAvg. Len.=128k, Mechanism=ALLMEM memory model2026.02 | 7.84 | 12.74 | |
| Qwen3-1.7B-MambaAvg. Len.=128k, Mechanism=Mamba-enhanced memory model2026.02 | 5.4 | 5.4 | |
| Qwen3-1.7B-SWA-SinksAvg. Len.=128k, Mechanism=Sliding-window attention, Sliding window length=8k, Attention sinks=1282026.02 | 4.16 | 4.48 | |
| Qwen3-1.7BAvg. Len.=128k, Mechanism=Full attention2026.02 | 2.43 | 2.42 |