ResearchDatasetsCombined (M3-Bench + Video-MME)Combined (M3-Bench + Video-MME)FollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsMultimodal Reasoning AverageCombined (M3-Bench + Video-MME-Long)46.8Average Score24