ResearchTasksWeb/tool tasksWeb/tool tasksFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedTool API 900 (test)CVT-RL86.9Success Rate3Jun 5, 2026AgentBench 700 (test)CVT-RL79.7Success Rate3Jun 5, 2026WebArena 245 (test)CVT-RL70.4Success Rate3Jun 5, 2026WebShop 1000 (test)CVT-RL81.2Success Rate3Jun 5, 2026