ResearchTasksJudge Agreement and Model RankingJudge Agreement and Model RankingFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedComparIAGemma4-26B-A4B0.156Kappa (κ)7Jun 12, 2026LMArena 140KGemma4-26B-A4B0.213Kappa (κ)7Jun 12, 2026