ScaleCUA: Scaling Open-Source Computer Use Agents with Cross-Platform Data

Zhaoyang Liu, Jingjing Xie, Zichen Ding, Zehao Li, Bowen Yang +25 more
9/18/2025
cs.CV

Official code: OpenGVLab/ScaleCUA (1,130 stars, Apache-2.0, last commit 2026-01-07).

Abstract

Vision-Language Models (VLMs) have enabled computer use agents (CUAs) that operate GUIs autonomously, showing great potential, yet progress is limited by the lack of large-scale, open-source computer use data and foundation models. In this work, we introduce ScaleCUA, a step toward scaling open-source CUAs. It offers a large-scale dataset spanning 6 operating systems and 3 task domains, built via a closed-loop pipeline uniting automated agents with human experts. Trained on this scaled-up data, ScaleCUA can operate seamlessly across platforms. Specifically, it delivers strong gains over baselines (+26.6 on WebArena-Lite-v2, +10.7 on ScreenSpot-Pro) and sets new state-of-the-art results (94.4% on MMBench-GUI L1-Hard, 60.6% on OSWorld-G, 47.4% on WebArena-Lite-v2). These findings underscore the power of data-driven scaling for general-purpose computer use agents. We will release data, models, and code to advance future research: https://github.com/OpenGVLab/ScaleCUA.

View on arXivView PDF

Code Implementations(1)

OpenGVLab/ScaleCUAOfficial100%
Apache-2.0

[ICLR 2026 Oral] ScaleCUA is the open-sourced computer use agents that can operate on cross-platform environments (Windows, macOS, Ubuntu, Android).

1,13080C++, CSSSep 9, 20259 months agoApache-2.0
computer-use-agentsdatagui-agentsmodelsonline-evaluation-suite+1 more

Similar Papers

OpenCUA: Open Foundations for Computer-Use Agents

92% match1 repo8392025

OS-Copilot: Towards Generalist Computer Agents with Self-Improvement

83% match1 repo1.8k2024

MAI-UI Technical Report: Real-World Centric Foundation GUI Agents

82% match1 repo2.3k2025

Windows Agent Arena: Evaluating Multi-Modal OS Agents at Scale

82% match1 repo8812026

OS-MAP: How Far Can Computer-Using Agents Go in Breadth and Depth?

82% match1 repo52025

AndroidLab: Training and Systematic Benchmarking of Android Autonomous Agents

81% match1 repo3252024

Cite this paper

@article{liu2025scalecua,
  title  = {ScaleCUA: Scaling Open-Source Computer Use Agents with Cross-Platform Data},
  author = {Zhaoyang Liu and Jingjing Xie and Zichen Ding and Zehao Li and Bowen Yang and Zhenyu Wu and Xuehui Wang and Qiushi Sun and Shi Liu and Weiyun Wang and Shenglong Ye and Qingyun Li and Xuan Dong and Yue Yu and Chenyu Lu and YunXiang Mo and Yao Yan and Zeyue Tian and Xiao Zhang and Yuan Huang and Yiqian Liu and Weijie Su and Gen Luo and Xiangyu Yue and Biqing Qi and Kai Chen and Bowen Zhou and Yu Qiao and Qifeng Chen and Wenhai Wang},
  year   = {2025},
  eprint = {2509.15221},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2509.15221v2}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2509.15221.svg)](https://codeofpaper.com/paper/2509.15221)

Discussion