ResearchTasksCollaborative ExecutionCollaborative ExecutionFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedAgentization Bench (Overall)Claude Code71.5Original Success Rate (Or.SR)4Apr 7, 2026Agentization Bench HardClaude Code44.4Original Success Rate (Or.SR)4Apr 7, 2026Agentization Bench (Medium)Claude Code67.2Or.SR4Apr 7, 2026Agentization Bench EasyCodex83.7Original Success Rate4Apr 7, 2026