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

MM-NIAH

Benchmarks

Task NameDataset NameSOTA ResultTrend
Long multimodal document understandingMM-NIAH
Overall Score46.1
7
Multi-modal Needle In A HaystackMM-NIAH 64K
Retrieval Score (Ret.)74.83
6
Long-context Multi-modal UnderstandingMM-NIAH 128K
Retrieval Score57.83
6
Showing 3 of 3 rows