VenusBench-Mobile - Hard GUI Browsing: leaderboard

Task completion of mobile GUI agents on VenusBench-Mobile (2026), user-intent Android tasks in emulated apps with screenshot observations and task-specific verification, on the hard gui browsing tasks (gathering and comparing information across many pages). Share of tasks completed, in percent; higher is better.

Metric: Task success rate (%) on the 34 hard GUI browsing (gathering and comparing information across many pages) tasks on VenusBench-Mobile, user-intent-driven Android tasks in emulated apps (screenshot-only observations, step-by-step GUI actions, task-specific verification functions with an MLLM checker where visual or semantic judgment is needed); higher is better. Source: arxiv.org. Saturation forecast: Rough model projection: around 2026. 7 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1UI-Venus-72B17.7
2Mobile-Agent-v3 + GUI-Owl-7B11.8
3Qwen 3 VL 30B A3B5.9#784
4MAI-UI-8B5.9
5GUI-Owl-7B2.9
6Qwen 3 VL 8B0#917
7UI-Venus-7B0

Interactive version: theaggregate.ai/benchmark?slug=venusbench-mobile-hard-gui-browsing · How It Works · Data refreshed daily, snapshot 2026-11-21.