VenusBench-Mobile - Conflict: 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 conflict tasks (instructions that contradict the screen state and need clarification). Share of tasks completed, in percent; higher is better.

Metric: Task success rate (%) on the 22 conflict (instructions that contradict the GUI state and need clarification) 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
1MAI-UI-8B13.6
2Mobile-Agent-v3 + GUI-Owl-7B9.1
3Qwen 3 VL 8B4.6#927
4Qwen 3 VL 30B A3B4.6#791
5UI-Venus-72B4.6
6UI-Venus-7B4.6
7GUI-Owl-7B0

Interactive version: theaggregate.ai/benchmark?slug=venusbench-mobile-conflict · How It Works · Data refreshed daily, snapshot 2026-11-22.