ResearchTasksmulti-turn interaction-based problem solvingFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedMINT-Bench 1.0 (test)LLAMA PRO - INSTRUCT11.76Code Generation Score5Feb 26, 2026