ResearchDatasetsMulti-node InfiniBand clusterFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsLLM Inference LatencyMulti-node InfiniBand cluster (2x8-GPU servers) inference performance (test)0.939Time To First Token (TTFT) Mean (s)4
LLM Inference LatencyMulti-node InfiniBand cluster (2x8-GPU servers) inference performance (test)0.939Time To First Token (TTFT) Mean (s)4