![[2602.09082] UI-Venus-1.5 Technical Report](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
[2602.09082] UI-Venus-1.5 Technical Report
Summary
The UI-Venus-1.5 Technical Report presents advancements in GUI agents, detailing a unified model that enhances task performance across various applications, showcasing state-of-the-art results.
Why It Matters
This report is significant as it addresses challenges in GUI automation, providing a robust solution that integrates multiple models into one. The advancements in training and performance metrics set a new benchmark for future research and applications in AI-driven GUI interactions.
Key Takeaways
- UI-Venus-1.5 introduces a unified GUI agent model for diverse applications.
- Key innovations include a Mid-Training stage and Online Reinforcement Learning.
- The model achieves state-of-the-art performance on multiple benchmarks.
- Demonstrates effective navigation in real-world Chinese mobile apps.
- Integrates domain-specific models into a cohesive framework.
Computer Science > Computer Vision and Pattern Recognition arXiv:2602.09082 (cs) [Submitted on 9 Feb 2026 (v1), last revised 24 Feb 2026 (this version, v2)] Title:UI-Venus-1.5 Technical Report Authors:Venus Team, Changlong Gao, Zhangxuan Gu, Yulin Liu, Xinyu Qiu, Shuheng Shen, Yue Wen, Tianyu Xia, Zhenyu Xu, Zhengwen Zeng, Beitong Zhou, Xingran Zhou, Weizhi Chen, Sunhao Dai, Jingya Dou, Yichen Gong, Yuan Guo, Zhenlin Guo, Feng Li, Qian Li, Jinzhen Lin, Yuqi Zhou, Linchao Zhu, Liang Chen, Zhenyu Guo, Changhua Meng, Weiqiang Wang View a PDF of the paper titled UI-Venus-1.5 Technical Report, by Venus Team and 26 other authors View PDF HTML (experimental) Abstract:GUI agents have emerged as a powerful paradigm for automating interactions in digital environments, yet achieving both broad generality and consistently strong task performance remains challenging. In this report, we present UI-Venus-1.5, a unified, end-to-end GUI Agent designed for robust real-world applications. The proposed model family comprises two dense variants (2B and 8B) and one mixture-of-experts variant (30B-A3B) to meet various downstream application scenarios. Compared to our previous version, UI-Venus-1.5 introduces three key technical advances: (1) a comprehensive Mid-Training stage leveraging 10 billion tokens across 30+ datasets to establish foundational GUI semantics; (2) Online Reinforcement Learning with full-trajectory rollouts, aligning training objectives with long-horizon, dynamic navigation i...
