ResearchDatasetsMatrix MultiplicationFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsLinear Kernel Performance256x256x1 Matrix Multiplication NVIDIA H800 GPU (test)5.54Latency (us)8Matrix MultiplicationMatrix Multiplication 3x3, 4x4, 5x54Circuit Depth2