ResearchTasksMultilingual Medical ReasoningMultilingual Medical ReasoningFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedCUREMED-BENCH (test)CURE-MED-Qwen2.5-32B94.96Consistency33Feb 26, 2026CUREMED-BENCH averaged across 13 languagesClaude 3 Haiku93.43Language Consistency5Feb 26, 2026