ResearchTasksGUI Task CompletionGUI Task CompletionFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedAndroidWorldMIRAGE-8B57.8Success Rate (SR)15Jun 30, 2026RISK-Bench (online)Qwen2.5-VL-72B88.7Completion Rate14Apr 14, 2026