ACE ROBOTICS

ACE ROBOTICS (大晓机器人) is presented here as a robotics-related commercial brand name. However, the only evidence supplied for this entry is an unrelated news article about Nvidia's reported acquisition of the open-source AI platform Hugging Face; that article contains no mention of ACE ROBOTICS, 大晓机器人, its products, ownership, headquarters, founding, or market activity. Because no brand-sense evidence was available, this dossier cannot responsibly describe the company's offerings, history, or positioning. The entry is therefore flagged for review, and the fields below are left empty rather than filled with guesses. If official-site or Wikipedia brand-sense material becomes available, this record should be rewritten with a full bilingual profile covering founding, headquarters, product lines, ownership, and market positioning.

• Unconfirmed

  ACEROBOTICS and Nanyang Technological University S-Lab release Mundo[2]大晓机器人与南洋理工大学 S-Lab 发布 Mundojd.zol.com.cn

  Reported: 2026-10-30

  ACEROBOTICS and Nanyang Technological University S-Lab released Mundo, a world model designed for building explorable, interactive environments.

• 2026-10-29

  Daxiao Robotics launches ACE-Brain-0.6 spatial intelligence API[3]大晓机器人推出 ACE-Brain-0.6 空间智能 APIchinaz.com

  Daxiao Robotics launched its ACE-Brain-0.6 embodied spatial intelligence API, which can recognize objects, locate them in three-dimensional space, and identify areas where people and objects can interact.

• 2026-10-19

  Daxiao Robotics announces ACE-Brain-0.6 embodied model[4]大晓机器人发布具身模型 ACE-Brain-0.6tech.ifeng.com

  Reported: 2026-10-19

  Daxiao Robotics announced its new ACE-Brain-0.6 vision-spatial-grounding-enhanced embodied model.

• 2026-08-28

  ACE Robotics and HKU release StreamPI for VLA temporal modeling[5]大晓机器人联合香港大学发布StreamPI163.com

  Reported: 2026-08-28

  ACE Robotics (大晓机器人), together with the University of Hong Kong, released StreamPI, a streaming multimodal temporal modeling method for Vision-Language-Action models that adds continuous temporal context to single-frame VLA decision making, with reported gains on LIBERO, CALVIN and real-robot tasks.

• Unconfirmed

  Daxiao Robotics and University of Hong Kong release StreamPI temporal modeling research[6]大晓机器人联合香港大学发布StreamPI时序建模研究成果finance.sina.com.cn

  Reported: 2026-08-28

  Daxiao Robotics, together with the University of Hong Kong, released StreamPI (Streaming Multimodal Temporal Modeling for Vision-Language-Action Models), a lightweight extension that gives existing VLA backbones continuous multimodal temporal context without adding extra model parameters. Experiments on real robots and benchmarks such as LIBERO and CALVIN reportedly improved temporal memory, fine spatial manipulation and long-horizon task performance.

1. 1斥资867亿元，80倍估值！英伟达收购开源AI平台抱抱脸-观察君-金融界finance.jrj.com.cn
2. 2大晓机器人与南洋理工大学 S-Lab 发布 Mundojd.zol.com.cn
3. 3大晓机器人推出 ACE-Brain-0.6 空间智能 APIchinaz.com
4. 4大晓机器人发布具身模型 ACE-Brain-0.6tech.ifeng.com
5. 5大晓机器人联合香港大学发布StreamPI163.com
6. 6大晓机器人联合香港大学发布StreamPI时序建模研究成果finance.sina.com.cn
7. 7大晓机器人联合南洋理工大学发布最新成果：Mundo，构建可自由探索、可物理交互的机器人世界模型ifanr.com
8. 8大晓机器人正式推出ACE-Brain-0.6 API_原创_科技频道首页_财经网 - CAIJING.COM.CNtech.caijing.com.cn
9. 9六项评测超越GPT6 Astra，大晓机器人发布具身新模型newsghexport.shobserver.cn