On 29 September, Hefei Lingfang Robotics disclosed a fresh Series A round, backed by market and industrial investors including the Wuxi Guolian Xizhou fund, Qingke Industrial Investment, Cowin Capital and Hefei Industrial Capital. The company was registered in Hefei’s Bahe district in January 2025, and the new round is its sixth in just nineteen months, a pace that says as much about the capital cycle as about the product.
A young team picks a vertical lane
Founder and chief executive Min Yuheng is a Tsinghua artificial-intelligence master’s graduate. The core team grew out of the university’s AI and Robotics lab and mixes PhD researchers with operators drawn from big internet companies, so technology and go-to-market sit in the same room. While most peers chase the spectacle of a prototype that walks, Lingfang chose early to prioritise deployment in narrow, paid scenarios rather than chase paper hardware specs.
The capital path is the interesting part. Local state funds such as Bahe Venture and the Hefei Seed Fund carried the company from prototype to product. As the hardware reached the market, industrial investors with auto-parts and tech-manufacturing backgrounds arrived, opening supply chains and real commercial orders. Local state capital solves the survival problem from zero to one. Industrial capital builds the bridge from one to many. That relay is becoming the textbook pattern for hard-tech startups in Anhui.
The technology bet
Lingfang’s in-house ZERITH Z1 biped carries 27 degrees of freedom, peak joint torque of 150 Nm and a load-to-weight ratio above 70 per cent, and its whole-body coordination controller lets it handle slopes, steps and uneven ground while grasping, carrying and interacting. The wheeled-arm H1 mates a wheeled base with a human upper body, reaching nearly 80 centimetres with a single arm, built for showfloor guidance, education and commercial service. A full-modal data loop uses VR teleoperation to teach the robot and capture force, touch, vision and joint data at once.
The company has already deployed across more than twenty landmark venues and reached single-month output in the low hundreds of units, covering exhibitions, tourism, education and services. The real threshold for embodied intelligence, the team argues, is not making a robot walk but making it finish tasks reliably in a noisy, changing world, and the data loop is what separates the leaders.
Fast money is not a business plan
Six rounds in nineteen months is rare even in China’s startup scene, and it is easy to read high frequency as a closed commercial loop. It is not. Lingfang deliberately avoids the consumer-home fantasy and attacks B-end service, exhibitions and education first, the routes where real orders and real interaction data arrive fastest. That is also the route where a young company can stay alive long enough to improve. Peer UBTECH took the same education-and-service door and still carries heavy R and D spend and persistent losses, the shared burden of the whole humanoid field.
Lingfang’s rise is one answer from a young local team: win vertical B-end orders, bank the data, then feed it back into the technology. Capital shortens the clock, but crossing the cycle still demands technology, product and commercial sense moving together.
More from the original report
Editor’s note: This is an adapted translation of the original OFweek Robotics report. It has been trimmed and restructured for readability for an international business audience.