Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

NIAH

Benchmarks

Task NameDataset NameSOTA ResultTrend
Information RetrievalNIAH (test)
Average Score99.6
59
Long-context RetrievalNIAH
Accuracy100
50
RetrievalNIAH seq=4096
Accuracy100
44
Multi-needle retrievalNIAH (M)
Accuracy (NIAH M)90.2
35
Needle-in-a-Haystack retrievalNIAH-Single-3 256K context
Accuracy94
23
Needle-in-a-Haystack retrievalNIAH-Single-3 128K context
Accuracy98.8
23
Needle-in-a-Haystack retrievalNIAH-Single-3 64K context
Accuracy100
23
Needle-in-a-Haystack retrievalNIAH-Single-2 32K context
Accuracy100
23
Needle-in-a-Haystack retrievalNIAH-Single-1 256K context
Accuracy100
23
Needle-in-a-Haystack retrievalNIAH-Single-1 128K context
Accuracy100
23
Needle-in-a-Haystack retrievalNIAH-Single-1 64K context
Accuracy100
23
Long-context RetrievalNIAH Single 3
Accuracy (1024)100
22
Long-context RetrievalNIAH 128k
Single Score24.4
20
Long-context RetrievalNIAH 64k
Single Score49.3
20
Long-context retrievalNIAH multivalue
Speedup4.1
20
Needle-In-A-Haystack RetrievalNIAH
NIAH Score100
14
Needle-in-a-Haystack retrievalNIAH Single 64K context 2
Accuracy100
13
Long Context RetrievalNIAH-Multi
Accuracy100
13
Long-context retrievalNIAH-Single-3 128K context
Accuracy90.6
12
Long-context retrievalNIAH-Single-3 64K context
Accuracy97
12
Long-context retrievalNIAH Single 3 32K context
Accuracy98.6
12
Long-context retrievalNIAH-Single-2 256K context
Accuracy93.4
12
Long-context retrievalNIAH-Single-1 256K
Accuracy100
12
Long-context retrievalNIAH-Single-1 128K
Accuracy100
12
Long-context retrievalNIAH-Single-1 32K
Accuracy100
12
Showing 25 of 58 rows