QCraft, a Chinese developer of autonomous driving solutions for global automakers, has secured $100 million in a Series D funding round as it focuses on the advancement of Level 4 self-driving technology and general physical AI.
The startup announced on Monday (March 23) the completion of a Series D round from a consortium of investors, including Ningbo Ninghai Xingtaihe Fund, Wonderland Capital, and the Liangxi Science and Innovation Industry Investment Fund Partnership, which is managed by Broad Vision Funds. An original equipment manufacturer (OEM) and automotive electronics components supplier also participated in the deal.
QCraft, which develops autonomous driving solutions spanning Level 2++ to Level 4, plans to invest the new financing in advancing frontier physical artificial intelligence (AI) R&D, with a focus on world models and reinforcement learning.
Its strategic focus on physical AI—artificial intelligence systems that combine advanced machine learning with sensors and actuators to perceive, reason about, and interact with the real world—comes amid a frenzy of dealmaking around the so-called “embodied AI” concept in China since 2025.
Unlike software-only AI, such as chatbots, physical AI acts on the physical world, enabling robots, vehicles, and smart systems to adapt to dynamic, unscripted environments. Such startups collectively raised over $1.9 billion through 43 deals in China over the first two months of 2026, according to DealStreetAsia’s data.
“2026 marks a critical inflection point in AI development. We are transitioning from ‘human-like’ intelligence to superhuman intelligence,” said Dr. James Yu, chairman and CEO of QCraft.
“Over the next five to ten years, the greatest opportunities in AI will emerge in the physical world—and that is precisely what makes autonomous driving so exciting. It is the best and most direct gateway into physical-world AI,” said Yu.
Besides physical AI, the Series D round will also allow QCraft to strengthen its organisational capabilities, global talent pipeline, and expansion internationally, the startup said.
The funding comes alongside several major milestones for QCraft. Its QPilot intelligent driving system surpassed one million vehicles in deployment across nearly 30 production models with close to 10 top-tier OEM partners. In 2026, the startup expects to add its technology and urban Navigate on Autopilot (NOA) capabilities to more than 50 new vehicle models.
In the Level 4 space, QCraft’s autonomous logistics vehicles have been commercially deployed in Jinhua, Wuhu, Ningbo, and other Chinese cities.
On the Robotaxi front, the startup plans to launch a new pilot programme in 2026 with full-scale deployment targeted for 2027.
Furthermore, its world model and reinforcement learning platform is also set for a near-term public debut.
Founded in Silicon Valley in 2019, QCraft has raised multiple funding rounds from both state and private capital investor including CICC Capital, ZGC Science City, Cuihu Capital, and TCL.
Venture investors at its early-stage seed and Series A rounds included IDG Capital, Vision Plus Capital, Tide Capital, Lenovo Capital and Incubator Group, YF Capital, and Meituan’s DragonBall Capital.
QCraft reportedly joined the confidential filing queue for a Hong Kong initial public offering (IPO), according to an IFR report in August 2025.
