ResearchTasksDownstream Language UnderstandingFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedOpen LLM Leaderboard zero-shotInheritune52.9ARCE6Feb 26, 2026OLMoE Downstream Evaluation Suite (MMLU, ARC-E, BoolQ, HellaSwag, PIQA, SciQ) 10B tokens (train)L2R (SIPS)27.6MMLU5Feb 26, 2026
OLMoE Downstream Evaluation Suite (MMLU, ARC-E, BoolQ, HellaSwag, PIQA, SciQ) 10B tokens (train)L2R (SIPS)27.6MMLU5Feb 26, 2026