GUANGZHOU, China, July 21, 2025 — WeRide (NASDAQ: WRD), a leading global company in autonomous driving technology, today unveiled its HPC 3.0 high-performance computing platform. This platform was co-developed with Lenovo (HKSE: 0992) and incorporates NVIDIA’s (NASDAQ: NVDA) latest DRIVE AGX Thor chips. The new HPC 3.0 platform is being introduced in WeRide’s newest-generation Robotaxi GXR, marking it as the world’s first mass-produced Level 4 (L4) autonomous vehicle to be built on NVIDIA DRIVE AGX Thor. Being fully automotive-grade, HPC 3.0 decreases the cost of the autonomous driving suite by 50%, thereby facilitating the extensive commercial deployment of the GXR.

WeRide’s HPC 3.0 platform includes a dual NVIDIA DRIVE AGX Thor setup that runs the safety-certified DriveOS. It is built upon Lenovo’s AD1 L4 autonomous driving domain controller, providing up to 2,000 TOPS of AI compute. This makes it the most powerful computing platform currently available for supporting L4 autonomy.

The HPC 3.0 platform, a collaboration between WeRide and Lenovo, features dual NVIDIA DRIVE AGX Thor chips and stands as the most robust computing platform for L4 autonomous driving.

In addition to its core enhancements, HPC 3.0 improves system integration, reducing mass production costs to one-quarter of HPC 2.0 and lowering autonomous driving suite costs by 50%. It also integrates essential modules like the Ethernet gateway, CAN gateway, inertial navigation, and collision detection, which leads to reduced production and maintenance expenses. Consequently, this lowers HPC 3.0’s total cost of ownership (TCO) by 84% throughout its operational lifecycle compared to its predecessor.

HPC 3.0 holds certifications from AEC-Q100, ISO 26262, and IATF 16949 standards. Its redundant architecture ensures the highest ASIL-D safety level, with a failure rate below 50 FIT (failures per billion hours of operation) and a MTBF (mean time between failures) ranging from 120,000 to 180,000 hours. Designed for 10 years or 300,000 km of use, it can operate in extreme temperatures from -40°C to 85°C and has passed tests for heat, shock, and corrosion. HPC 3.0 also complies fully with global VOCs environmental standards, making it suitable for deployment across the Middle East, Southeast Asia, Europe, and other international markets, supporting WeRide’s objective to expand its Robotaxi fleet globally.

“Our close collaboration with Lenovo and NVIDIA represents a significant advancement in computing power, system architecture, and cost efficiency. By integrating the HPC 3.0 platform into our Robotaxi GXR, we enhance vehicle reliability and responsiveness while substantially decreasing deployment costs. Going forward, we intend to implement this platform across more of our L4 autonomous vehicles—including the Robobus and Robosweeper—to bring intelligent, accessible mobility to more cities and customers worldwide,” stated Tony Han, Founder and CEO of WeRide.

“We are excited by the successful deployment of NVIDIA DRIVE AGX Thor on WeRide’s Robotaxi GXR, which marks a significant step toward bringing L4 autonomous vehicles to market at scale. Our close collaboration with WeRide on the AD1 domain controller has been crucial in accelerating this innovation. Lenovo is committed to leveraging our advanced computing power and collaborating with industry partners to drive the global commercialization of autonomous driving solutions,” commented Donny Tang, Vice President and Head of Lenovo Vehicle Computing, Lenovo.

NVIDIA has served as a strategic investor in WeRide since 2017 through the NVIDIA Inception program. Ali Kani, Vice President of Automotive at NVIDIA, also extended his congratulations to WeRide.

“Robotaxis are transforming urban mobility, and it is thrilling to observe the progress WeRide and Lenovo have made in deploying a Level 4 autonomous driving system built on NVIDIA accelerated compute and DriveOS,” said Ali Kani, Vice President of Automotive, NVIDIA. “Their accomplishment signifies a major milestone in making safe, scalable, and efficient autonomous transportation a reality.”

NVIDIA DRIVE AGX Thor, the successor to NVIDIA’s DRIVE AGX Orin, is constructed on the NVIDIA Blackwell architecture and is optimized for the most demanding processing workloads, including generative AI, vision language models, and large language models. Its streamlined architecture improves generalization, reduces latency, and enhances safety by utilizing powerful NVIDIA accelerated computing to run WeRide’s proprietary AV software stack.

As the world’s first publicly listed Robotaxi company, WeRide has safely operated Robotaxis on public roads for over 2,000 days. The company continues to expand its technology through global partnerships, providing safe, efficient, and cost-effective autonomous mobility.

About WeRide
WeRide is a global leader and pioneer in the autonomous driving industry, recognized as the first publicly traded Robotaxi company. Our autonomous vehicles have been tested or operated in more than 30 cities across 10 countries. We are also the first and only technology company whose products have obtained autonomous driving permits in five markets: China, the UAE, Singapore, France, and the US. Powered by the intelligent, versatile, cost-effective, and highly adaptable WeRide One platform, WeRide offers autonomous driving products and services ranging from L2 to L4, addressing transportation needs in the mobility, logistics, and sanitation sectors. WeRide was featured in Fortune Magazine’s 2024 “The Future 50” list.

Media Contact
pr@weride.ai