ResearchTasksEnd-to-end throughputEnd-to-end throughputFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedLLaMA-2-7B-ChatCPE449Throughput (tokens/sec)60Feb 26, 2026Cloud Inference WorkloadsSpeContext824.22Throughput (tokens/s)32Feb 26, 2026