| title | CoworkOS:Scaling Agent Coworkers for Real-world Automation |
|---|---|
| authors | Jiajie Jin, Guanting Dong, Chengjun Pan, Xin Deng, Zhuohui Sheng, Guoqiang Wang, Tao Gui, Yutao Zhu, Yujia Chen, Shihong Yi, Binze Hu, Zhixiong Yang, Zhichao Sun, Rui Wang, Han Han, Mingyu Ma, Junhao Huang, Xiaotao Wang, Huicheng Jiang, Yuyang Hu, Xinyu Yang, Wanghan Xu, Yajie Yang, Zhecan Wang, Honglin Guo, Xiaoyu Xing, Xiaoqian Zhang, Yulun Zhu, Xiaoxi Li, Jiahang Lin, Shaofan Liu, Shihan Dou, Zhiheng Xi, Qi Zhang, Yu-Gang Jiang, See-Kiong Ng, Yang Zhang, Tat-Seng Chua, Ji-Rong Wen, Zhicheng Dou |
| date | 2026-11-02 |
| venue | arXiv (preprint, not peer reviewed) |
| id | arXiv:2611.03178 |
| affiliation | Renmin University of China |
| upvotes | 88 (Hugging Face Daily Papers, 2026-11-04) |
| links | abs pdf hf code |
| cite | |
| abstract | Rapid advances in cowork agents are shifting human-AI collaboration toward real-world automation, delivering software, designs, and other artifacts. However, such productivity work demands both high-quality delivery on complex productivity tasks and continued monitoring and decision-making over long horizons, a combination that remains an open challenge. We introduce CoworkOS, a general-purpose harness for long-horizon work that separates task strategy from execution management: the model decides actions, delegation, and acceptance, while the runtime maintains state and feedback. Its three-layer architecture enables direct execution or delegation to Subagents, synchronously or asynchronously, under stable responsibilities. Through a shared application layer, humans and agents inspect and revise the same evolving documents, presentations, CAD models, and Blender scenes, so human-in-the-loop participation can enter both through conversation and through direct modification of the work itself. To sustain continuity across agents and sessions, CoworkOS further introduces hierarchical workspaces and persistent semantic memory, so working files, key decisions, and long-running Job logs remain shareable, reusable, and resumable after interruption. Across 12 benchmarks and 6 categories of real-world applications, CoworkOS proves highly competitive with frontier agent applications such as Codex and DeepSeek Harness. As a sustained-work case study, CoworkOS uses GLM-5.3 as its backbone to conduct RSI-driven automated post-training on Qwen3.6-35B-A3B. Over five rounds, the model's average score across eight benchmarks rose from 33.19 to 42.95 (+9.76). Looking ahead, we believe CoworkOS opens a meaningful path toward general cowork-agent intelligence. |
bibtex
@misc{jin2026coworkos,
title={CoworkOS:Scaling Agent Coworkers for Real-world Automation},
author={Jiajie Jin and Guanting Dong and Chengjun Pan and Xin Deng and Zhuohui Sheng and Guoqiang Wang and Tao Gui and Yutao Zhu and Yujia Chen and Shihong Yi and Binze Hu and Zhixiong Yang and Zhichao Sun and Rui Wang and Han Han and Mingyu Ma and Junhao Huang and Xiaotao Wang and Huicheng Jiang and Yuyang Hu and Xinyu Yang and Wanghan Xu and Yajie Yang and Zhecan Wang and Honglin Guo and Xiaoyu Xing and Xiaoqian Zhang and Yulun Zhu and Xiaoxi Li and Jiahang Lin and Shaofan Liu and Shihan Dou and Zhiheng Xi and Qi Zhang and Yu-Gang Jiang and See-Kiong Ng and Yang Zhang and Tat-Seng Chua and Ji-Rong Wen and Zhicheng Dou},
year={2026},
eprint={2611.03178},
archivePrefix={arXiv},
url={https://arxiv.org/abs/2611.03178}
}