ResearchTasksComprehensive Multimodal EvaluationFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedSEED-Bench ImageSAIL-VL277.3Accuracy10Feb 26, 2026Multimodal Evaluation Suite CompositeEVE (Ours-8B-iter4)68.7Overall Score5Apr 21, 2026