ResearchTasksMedical Visual Question Answering / Tool-useMedical Visual Question Answering / Tool-useFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedChestAgentBench OOD (test)Gemini-3-flash76.2Accuracy7Mar 11, 2026