ResearchTasksAgentic Phone UseAgentic Phone UseFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updated150-task Real-phone Human Evaluation Single-AppPhoneBuddy-4B-Real+Mock62Task Success Rate14Jun 23, 2026AndroidWorldPhoneBuddy-4B-Real+Mock83.2Task Success Rate7Jun 23, 2026150-task Real-phone Human Evaluation Cross-AppGemini 3.1 Pro48Task Success Rate7Jun 23, 2026
150-task Real-phone Human Evaluation Single-AppPhoneBuddy-4B-Real+Mock62Task Success Rate14Jun 23, 2026