ResearchTasksMultilingual Multiple-Choice ReasoningFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedGlobal PIQA 116 languages 1.0 (test)Qwen3.5-4B79.31Accuracy6Mar 13, 2026INCLUDE 44 languages 1.0 (test)Qwen3.5-4B56.9Average Accuracy6Mar 13, 2026Global MMLU 42 languages 1.0 (test)Qwen3.5-4B54.8Average Accuracy6Mar 13, 2026