ResearchTasksGeneral Intelligence EvaluationFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedAGI-EvalQwen7B61.2Accuracy10May 20, 2026AGI Eval EnglishQwen 3 VL 32B Think92.2Score8Feb 26, 2026AGIEval (test)Full FT27AGIEval (3-shot)8Feb 26, 2026AGIEval GSA-SFT72Accuracy4Feb 26, 2026