Developing autonomous agents that effectively interact with Graphic User Interfaces (GUIs) remains a challenging open problem, especially for small on-device models. In this paper, we present Ferret-UI Lite, a compact, end-to-end GUI agent that operates across diverse platforms, including mobile, web, and desktop. Utilizing techniques optimized for developing small models, we build our 3B Ferret-UI Lite agent through curating a diverse GUI data mixture from real and synthetic sources, strengthening inference-time performance through chain-of-thought reasoning and visual tool-use, and reinforcement learning with designed rewards. Ferret-UI Lite achieves competitive performance with other small-scale GUI agents. In GUI grounding, Ferret-UI Lite attains scores of 91.6%, 53.3%, and 61.2% on the ScreenSpot-V2, ScreenSpot-Pro, and OSWorld-G benchmarks, respectively. For GUI navigation, Ferret-UI Lite achieves success rates of 28.0% on AndroidWorld and 19.8% on OSWorld. We share our methods and lessons learned from developing compact, on-device GUI agents.
@article{arxiv.2509.26539,
title = {Ferret-UI Lite: Lessons from Building Small On-Device GUI Agents},
author = {Zhen Yang and Zi-Yi Dou and Di Feng and Forrest Huang and Anh Nguyen and Keen You and Omar Attia and Yuhao Yang and Michael Feng and Haotian Zhang and Ram Ramrakhya and Chao Jia and Jeffrey Nichols and Alexander Toshev and Yinfei Yang and Zhe Gan},
journal= {arXiv preprint arXiv:2509.26539},
year = {2025}
}