Beijing Rimu Robot / 日冕开物
About
About Beijing Rimu Robot / 日冕开物
Beijing Rimu Robot develops model and data infrastructure for teaching industrial robots to perform variable physical tasks. Its LaMPA stack combines world modelling, robot-policy learning and reward-model development, allowing task behaviour to be trained from demonstrations and evaluated against learned objectives. The company is targeting work that requires adaptation across changing objects, scenes and production conditions.
The wider workflow includes data collection, simulation, model training and deployment agents that connect learned policies with industrial equipment. Initial activity centres on manipulation and process tasks where fixed programming creates high integration effort. Rimu Robot remains an early-stage company building its technical team and conducting application development with industrial partners.
Articles
Funding
- 2026-07-17 - Two seed rounds - Beijing Rimu Robot raises two seed rounds
Beijing Rimu Robot completed two seed rounds totaling several hundred million yuan. - source: 36kr.com
Investors
- Baidu Ventures - Participant - VC - 1 funding event - first seen 2026-07-17
- Dingfeng - Participant - Unknown - 1 funding event - first seen 2026-07-17
- Wanlin International - Participant - Angel / Individual - 1 funding event - first seen 2026-07-17
- WoYan Capital - Participant - VC - 1 funding event - first seen 2026-07-17
- Wuyuefeng - Participant - Unknown - 1 funding event - first seen 2026-07-17
- Yuantu Future - Participant - Angel / Individual - 1 funding event - first seen 2026-07-17
Timeline
Events
- 2026-07-17 - Funding - Beijing Rimu Robot raises two seed rounds
Participating investors include Dingfeng, Yuantu Future, Baidu Ventures, WoYan Capital, Wuyuefeng and Wanlin International. The proceeds support LaMPA model development, reinforcement learning, data feedback loops and product delivery. LaMPA combines environment, robot-self and experience representations; its World Reward Model provides automated feedback for reinforcement learning.
Source