ResearchTasksGeneral Capability Evaluation (Held-out Benchmarks)FollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedOlmoBaseEval LBPP BBH MMLU Pro MC Deepmind Math (HeldOut)Seed 36B42.6LBPP Score24Feb 26, 2026