ResearchTasksGraduate-level Science QAGraduate-level Science QAFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedGPQA MainRaw CoT73.2Accuracy20Jun 30, 2026GPQA DDRE5,318Tokens Used for Thinking15Jun 17, 2026GPQAParallel-Synthesis52.02Accuracy13Jun 15, 2026