Gasgoo Munich- Embodied intelligence startup Noematrix has recently closed a new funding round worth hundreds of millions of yuan.
With this capital, Noematrix plans to push forward the development and iteration of embodied intelligence large models—focusing on strong generalization and autonomous decision-making—to speed up deployment in real-world settings like hotel laundry and home organization.
The company boasts a full-stack, closed-loop capability that spans everything from proprietary model development and real-world data collection to in-house hardware creation—feeding deployment insights back into model optimization.
At the model level, Noematrix is committed to a self-developed general embodied intelligence path. Leveraging massive real-world data for pre-training and hybrid force-position post-training, the company continues to iterate the Noematrix Brain—a general embodied intelligence engine for physical environments. This system equips robots with a complete decision loop, covering instruction understanding, task planning, environmental perception, and execution feedback.
Image source: Noematrix
Currently, Noematrix Brain 2.0 has evolved embodied intelligence from a "demonstrable capability" to a "system capable of sustained operation in real-world environments." Built around an embodied brain architecture, a real-world data loop, and multi-scenario task execution, the system establishes an intelligent decision-making framework for the physical world—enabling robots to coordinate complex tasks and operate stably within dynamic, uncertain conditions.
In the near term, the company plans to officially launch its next-generation proprietary embodied intelligence world model.
On the data front, Noematrix has partnered with Wuxi Data Group and others to launch a city-wide initiative dubbed the "Thousand Enterprises, Million Hours" consortium for high-quality embodied intelligence datasets. The group has officially released the initial results of this first phase.
Moving forward, the partners will leverage Wuxi Data Group's industrial resources and scenario governance capabilities, combined with Noematrix's technical expertise in frontier models and embodied intelligence, to deepen industrial dataset construction. The goal is to drive the deployment of embodied intelligence on actual production lines.
Regarding deployment, Noematrix robots are already operating in batches across pharmacies and hotels. The company's "embedded upgrade" approach requires no structural changes to existing shelving and needs just 2.5 square meters of space. It integrates directly with current store order systems to run stably within existing aisles—establishing a replicable, scalable benchmark for the commercial application of embodied intelligence.
About Seeds Discovery:
Gasgoo's "Seeds Discovery" column is designed to create a service platform connecting startups, ecosystem partners, investors, and local governments—deeply empowering the industry's upstream and downstream. Since its inception, the column has focused on uncovering standout companies, technologies, and business models that offer inspiration and leadership amid the wave of intelligent transformation, thereby driving industrial innovation. According to Gasgoo data, nearly every startup featured in Seeds Discovery has successfully connected with ecosystem resources across the supply chain.
