ResearchTasksAggregate Multimodal PerformanceAggregate Multimodal PerformanceFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedMultimodal Evaluation SummaryQwen3-VL-2B74.89Average Accuracy16May 19, 2026AverageQwen2.5-VL-72B + CARES80Score10Mar 23, 2026