| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Information Retrieval | NIAH (test) | Average Score99.6 | 59 | |
| Long-context Retrieval | NIAH | Accuracy100 | 50 | |
| Retrieval | NIAH seq=4096 | Accuracy100 | 44 | |
| Multi-needle retrieval | NIAH (M) | Accuracy (NIAH M)90.2 | 35 | |
| Needle-in-a-Haystack retrieval | NIAH-Single-3 256K context | Accuracy94 | 23 | |
| Needle-in-a-Haystack retrieval | NIAH-Single-3 128K context | Accuracy98.8 | 23 | |
| Needle-in-a-Haystack retrieval | NIAH-Single-3 64K context | Accuracy100 | 23 | |
| Needle-in-a-Haystack retrieval | NIAH-Single-2 32K context | Accuracy100 | 23 | |
| Needle-in-a-Haystack retrieval | NIAH-Single-1 256K context | Accuracy100 | 23 | |
| Needle-in-a-Haystack retrieval | NIAH-Single-1 128K context | Accuracy100 | 23 | |
| Needle-in-a-Haystack retrieval | NIAH-Single-1 64K context | Accuracy100 | 23 | |
| Long-context Retrieval | NIAH Single 3 | Accuracy (1024)100 | 22 | |
| Long-context Retrieval | NIAH 128k | Single Score24.4 | 20 | |
| Long-context Retrieval | NIAH 64k | Single Score49.3 | 20 | |
| Long-context retrieval | NIAH multivalue | Speedup4.1 | 20 | |
| Needle-In-A-Haystack Retrieval | NIAH | NIAH Score100 | 14 | |
| Needle-in-a-Haystack retrieval | NIAH Single 64K context 2 | Accuracy100 | 13 | |
| Long Context Retrieval | NIAH-Multi | Accuracy100 | 13 | |
| Long-context retrieval | NIAH-Single-3 128K context | Accuracy90.6 | 12 | |
| Long-context retrieval | NIAH-Single-3 64K context | Accuracy97 | 12 | |
| Long-context retrieval | NIAH Single 3 32K context | Accuracy98.6 | 12 | |
| Long-context retrieval | NIAH-Single-2 256K context | Accuracy93.4 | 12 | |
| Long-context retrieval | NIAH-Single-1 256K | Accuracy100 | 12 | |
| Long-context retrieval | NIAH-Single-1 128K | Accuracy100 | 12 | |
| Long-context retrieval | NIAH-Single-1 32K | Accuracy100 | 12 |