en.Wedoany.com Reported - Recently, Guanglun Intelligence, a Chinese physical AI infrastructure company, completed a new round of 1 billion yuan strategic financing. Investors in this round include government funds such as Zhongguancun Science City Fund, Sichuan Development Science and Innovation Fund, and Shandong Development Science and Technology Venture Capital, as well as industrial capital and financial investment institutions like Giant Network, Yusys Technologies, Baotong Technology, Zhongke Industry Investment, and Liangtu Zhice. Existing shareholders, including Jiantou Investment, 37 Interactive Entertainment, and Semir Investment, continued to follow the investment.
This round of financing will be primarily used to continuously increase investment in core technology research and development for physical AI data and evaluation infrastructure. Guanglun Intelligence will further improve its product system for robot learning, capability evaluation, and real-world scenario deployment, expand the construction of high-quality human behavior data, simulation synthetic data, and industrial-grade evaluation capabilities, and jointly promote the development of an open ecosystem with industry partners.
Guanglun Intelligence focuses on data, simulation, and evaluation infrastructure in the era of physical AI. Unlike traditional software AI, physical AI requires robots to perform perception, decision-making, and action execution in real environments. Data comes not only from text and images but also from human behavior, object interactions, spatial relationships, motion trajectories, and scene feedback. High-quality physical data and reproducible simulation environments are becoming an important foundation for improving robot capabilities.
Robot learning has high requirements for data quality. Real-world tasks often involve long-tail scenarios, such as object deformation, contact friction, occlusion changes, complex lighting, personnel flow, and non-standard operations. Relying solely on a small amount of real collected data results in high model training costs, limited coverage, and difficulty in quickly completing safety verification. The combination of simulation synthetic data and human behavior data helps improve the generalization ability of robots in complex scenarios.
Evaluation infrastructure is also a key link in the deployment of embodied intelligence. Whether a robot has stable capabilities for handling, recognition, grasping, inspection, collaboration, and anomaly handling cannot be judged solely by laboratory demonstrations; it also needs to be verified in standardized tasks, industrial-grade scenarios, and repeatable testing environments. Guanglun Intelligence's continued investment in evaluation capability construction in this round indicates that capital is focusing on the foundational link of embodied intelligence moving from "model training" to "capability verification" and "scenario delivery."
From an application perspective, Guanglun Intelligence's product system will continue to target robot learning, real-world scenario deployment, and collaboration with industry partners. Scenarios such as industry, retail, logistics, manufacturing, and services have different requirements for robot capabilities, but they all share the need for scalable data supply, verifiable simulation environments, and quantifiable capability evaluation systems. The more complete the data and evaluation capabilities, the easier it is for robot companies to reduce trial-and-error costs and shorten product iteration cycles.
This round of financing also reflects the rising investment value of physical AI infrastructure. In the past, the embodied intelligence track focused more on robot bodies, joints, motors, controllers, and large models. Now, the market is beginning to pay further attention to underlying data, simulation, evaluation, and Sim2Real conversion capabilities. For the robot industry, only when data, models, simulation, and real-world scenarios form a closed loop can robots move from single-point demonstrations to large-scale deployment.
However, physical AI infrastructure is still in a rapid evolution stage. The cost of high-quality data collection, simulation realism, standardization of evaluation criteria, cross-scenario reuse capabilities, and commercial delivery efficiency will all affect the subsequent development of related companies. After completing this new round of strategic financing, whether Guanglun Intelligence can transform its data and evaluation capabilities into stable customer projects and industry-standard products will be key to testing its competitiveness.
Subsequent observation focuses will be on Guanglun Intelligence's core technology research and development progress, the improvement of its robot learning product system, the implementation effect of its industrial-grade evaluation platform, the depth of collaboration with industry partners, and the application scale of its data and simulation capabilities in real robot projects. As the embodied intelligence industry continues to heat up, physical AI data and evaluation infrastructure is expected to become an important support layer for robot commercialization.
This article is compiled by Wedoany. All AI citations must indicate the source as "Wedoany". If there is any infringement or other issues, please notify us promptly, and we will modify or delete it accordingly. Email: news@wedoany.com
