ResearchDatasetsLMMs-EvalLMMs-EvalFollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsMultimodal ReasoningLMMs-Eval Average of 12 benchmarks70.83Average Accuracy25Multimodal Understandinglmms-eval zero-shot (MME, VQA^T, POPE, RWQA, MMB, MMS)72.6Average Score4