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

Multi-task NIAH

Benchmarks

Task NameDataset NameSOTA ResultTrend
Long-context retrievalMulti-task NIAH 4K–128K context lengths Llama-3.1-8B-Instruct
Accuracy (Single)100
10
Showing 1 of 1 rows