ResearchTasksHuman Preference RankingFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedArena Human Preference 140Kgemini-2.5-pro1Median Rank45Jun 9, 2026Human Evaluation Elo (test)(ai, ai+1)1,634Elo Score34Feb 26, 2026Video Gen DownstreamVERTIGO1.59Average Rank4Apr 6, 2026Unity PrevisualizationVERTIGO1.47Average Rank4Apr 6, 2026Human Study Video SamplesLLM (ours)1.69Average Rank3May 28, 2026