Reproduce “Mage-VL-Cowork: A Full Training Recipe to Learn Hybrid Computer Use From Evolved Harnesses” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Xiaoyi Zhang, Haoqing Wang, Xinjie Zhang, Xiang Ming, Xing Wang, Senqiao Yang, Kaichen Zhang, Zhaoyang Jia and 14 more
Mage-VL-Cowork trains a 4B desktop agent to choose between screenshot-based GUI actions and code in a persistent terminal. For a showcased Calc task, it edits a workbook in 6 steps and 0:39, versus 39 steps and 9:56 for GUI-only Qwen3.6-27B. It scores 66.0% on OSWorld-Verified’s 361 held-out tasks under 50 steps; this is not a controlled speed comparison.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.