ResearchDatasetsStreaming-EvalFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsNarrationStreaming-Eval70Win Rate vs. Claude Opus 4.6 Thinking8Question AnsweringStreaming-Eval QA 1.0 (test)54.2Backward Score8