Mobile-Agent-v3: Fundamental Agents for GUI Automation

TL;DR

This paper presents GUI-Owl, a foundational GUI agent model, and Mobile-Agent-v3, a new framework that significantly advances open-source GUI automation across multiple platforms through innovative environment infrastructure, decision-making capabilities, and scalable reinforcement learning.

Contribution

Introduction of Mobile-Agent-v3, a comprehensive GUI agent framework that improves performance and incorporates novel environment infrastructure, decision-making, and reinforcement learning techniques.

Findings

GUI-Owl achieves state-of-the-art results on GUI benchmarks.

Mobile-Agent-v3 outperforms previous models with 73.3 on AndroidWorld.

The framework supports diverse platforms and scalable RL methods.

Abstract

This paper introduces GUI-Owl, a foundational GUI agent model that achieves state-of-the-art performance among open-source end-to-end models on ten GUI benchmarks across desktop and mobile environments, covering grounding, question answering, planning, decision-making, and procedural knowledge. GUI-Owl-7B achieves 66.4 on AndroidWorld and 29.4 on OSWorld. Building on this, we propose Mobile-Agent-v3, a general-purpose GUI agent framework that further improves performance to 73.3 on AndroidWorld and 37.7 on OSWorld, setting a new state-of-the-art for open-source GUI agent frameworks. GUI-Owl incorporates three key innovations: (1) Large-scale Environment Infrastructure: a cloud-based virtual environment spanning Android, Ubuntu, macOS, and Windows, enabling our Self-Evolving GUI Trajectory Production framework. This generates high-quality interaction data via automated query generation and correctness validation, leveraging GUI-Owl to refine trajectories iteratively, forming a self-improving loop. It supports diverse data pipelines and reduces manual annotation. (2) Diverse Foundational Agent Capabilities: by integrating UI grounding, planning, action semantics, and reasoning patterns, GUI-Owl supports end-to-end decision-making and can act as a modular component in multi-agent systems. (3) Scalable Environment RL: we develop a scalable reinforcement learning framework with fully asynchronous training for real-world alignment. We also introduce Trajectory-aware Relative Policy Optimization (TRPO) for online RL, achieving 34.9 on OSWorld. GUI-Owl and Mobile-Agent-v3 are open-sourced at https://github.com/X-PLUG/MobileAgent.