ResearchTasksMath and ReasoningMath and ReasoningFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedMathVista miniQwen3.5-Omni82.9Overall Score26May 12, 2026GSM8KBFPO83.85Accuracy20Feb 26, 2026LogicVistaBee-8B61.3Score6Mar 10, 2026DynaMath worstBee-8B41.3Score6Mar 10, 2026