ResearchTasksVerifiable JudgingVerifiable JudgingFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedVerifyBench HardOpenAI/GPT-o188.8Accuracy19Jun 12, 2026VerifyBench (full)OpenAI/GPT-o195.7Accuracy19Jun 12, 2026