- Physical AI, aerospace and world model startups secured fresh funding as investors doubled down on next-generation computing infrastructure
- The latest rounds backed physics engines, satellite propulsion systems and AI models designed to understand and simulate the physical world
Fysics closes Series A1 round to expand physical AI platform
Shanghai-based physical AI company Fysics (飞捷科思) has completed a Series A1 funding round worth hundreds of millions of yuan.
The round was jointly led by existing investor TedaVC together with Oceanpine and Fudan Capital. New investors included Addor Capital, Pre IPO Capital, Hardstone Capital, Huirong Fund, Fortune Prosperity Capital and Xiangjiang State Capital Investment.
Existing backers such as Yunqi Partners, Silicon Harbour, Zhongying Capital and Conswall Ventures increased their investments.
The company said the proceeds will support further development of its full-stack technology platform for embodied AI and advanced manufacturing applications.
Founded by a team led by Zhang Lihua (张立华), a professor at Fudan University and one of the founding architects of Nvidia’s PhysX simulation engine, Fysics has developed five core products.
They include the Fysics differentiable physics engine, MoziSim embodied AI simulation platform, OmniFysics multimodal foundation model, Fysiverse world model and FysiData data platform.
Guoke Xingqing Spacetech completes new funding round
Hangzhou-based space propulsion developer Guoke Xingqing Spacetech (国科星擎) has completed a fresh funding round worth tens of millions of yuan.
Investors included Hangzhou STI Group’s Runmiao Fund, Shanghai STVC Group, Huiyan Investment and Gongqingcheng Xuri Venture Capital.
The company plans to use the proceeds to expand research and development, scale production capacity and strengthen market expansion efforts.
Incubated at the University of Chinese Academy of Sciences’ Hangzhou Institute for Advanced Study, the company is the only private Chinese business to achieve large-scale in-orbit verification and commercial deployment of krypton-fueled Hall thrusters.
Its products cover thrust levels ranging from hundreds of micronewtons to 300 millinewtons and can be deployed on platforms ranging from CubeSats to medium- and high-orbit satellites.
Awomo raises nearly $15 million in seed and angel financing
Hangzhou-based concept world model developer Awomo (西湖数智) has completed seed and angel financing rounds totaling more than 100 million yuan ($14.8 million).
Investors included Innoangel Fund, Orient Renaissance Capital, Zhengxuan Investment, Tianqi Capital, Westlake Innovation Capital and Golden Ant Capital.
The startup was incubated at Westlake University‘s Autonomous Intelligence Laboratory. Founder Yu Kaicheng (于开丞) is a distinguished researcher at Westlake University and previously worked at Alibaba’s DAMO Academy.
Unlike conventional simulation approaches based on pixel-level modeling, Awomo’s technology relies on latent-state transition models to build a general-purpose physical world engine for embodied AI, industrial automation and complex decision-making scenarios.
The company said the funds will be used to build large-scale datasets of physical interactions, expand computing resources and improve long-horizon planning capabilities.
