ResearchTasksGeneral AI Assistant Task ExecutionGeneral AI Assistant Task ExecutionFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedGAIA Round 1KIMI-K225Level 1 Evo. Pass@18Mar 25, 2026GAIA Round 2Deepseek-v3.120Level 1 Evo. Pass@14Mar 25, 2026