Shanghai (Gasgoo)- Chinese autonomous driving startup Pony.ai unveiled its seventh-generation Robotaxi, built on the Toyota bZ4X, at the 2025 International Automotive & Supply Chain Expo (Hong Kong) ("2025 Hong Kong Auto Expo") held from June 12 to 15 in the AsiaWorld-Expo. This marked the model's debut in the city, showcasing the company's latest advancements in autonomous driving hardware and software.
Specifically engineered for large-scale Level 4 autonomous vehicle deployment, Pony.ai's Gen-7 Robotaxi utilizes 100% automotive-grade components, and boasts a 70% cost reduction in its autonomous driving kits compared to previous generations. Designed for adaptability, it supports integration across multiple vehicle models.
Equipped with 34 sensors spanning six categories—including 9 LiDAR units, 14 cameras, and 4 millimeter-wave radars — the Gen-7 Robotaxi offers comprehensive detection coverage from full 360-degree blind spot monitoring to identifying objects up to 650 meters away. It is capable of operating in a variety of urban and intercity environments, including highways, ring roads, airports, and railway stations.
A standout feature of the Gen-7 Robotaxi is its autonomous driving domain controller, which is the world's first Level 4 kind to house four NVIDIA Orin-X chips. Delivering a total computing power of 1,016 TOPS, the system is designed with a lifecycle of 10 years or 600,000 kilometers. Further software optimizations have tripled its computational efficiency compared to previous generations.
Also powered by Pony.ai's seventh-generation autonomous driving system, the GAC AION's second-generation AION V has recently secured a road test license of intelligent connected vehicle. The vehicle's public road testing is now underway in Guangzhou and Shenzhen.
As the only company in China authorized to operate autonomous ride-hailing services in all four of the nation's Tier-1 cities, namely, Beijing, Shanghai, Guangzhou, and Shenzhen, Pony.ai continues to expand its operational network across these cities. The addition of the Gen-7 Robotaxi is expected to significantly expand the company's fleet, which is projected to exceed 1,000 vehicles by the end of 2025.
Hong Kong, one of China's premier global hubs, plays a strategic role in international connectivity and finance. Earlier this year, Pony.ai showcased its sixth-generation Robotaxi at the Hong Kong International Airport and plans to gradually roll out ride-hailing services throughout the city.
Looking ahead, integrating into the Hong Kong market is expected to facilitate greater connectivity for intelligent connected vehicles across the Guangdong-Hong Kong-Macao Greater Bay Area ("GBA"). Under the Framework Agreement on Mutual Recognition and Cooperation for Pilot Demonstration of Intelligent Connected Vehicles signed earlier this year, Guangzhou's Nansha, Zhuhai's Hengqin, and Shenzhen's Bao'an and Qianhai districts are leading efforts to establish cross-regional systems for qualification recognition, road access, and regulatory coordination. Pony.ai is among the first autonomous driving companies to obtain cross-regional mutual recognition status under this framework.
Successful deployment in Hong Kong would allow Pony.ai's Robotaxis to traverse multiple zones within the GBA — including airports and train stations — enabling seamless intercity autonomous transportation and enhancing mobility for residents with safe, reliable and convenient self-driving services.
While strengthening its domestic presence, Pony.ai is also accelerating global expansion. The company has launched autonomous driving operations in other countries and regions such as South Korea, Luxembourg, and the Middle East, thus exporting its technologies and products. It has also formed strategic partnerships with Uber, Singapore's ComfortDelGro, and Luxembourg's Emile Weber, aiming to bring Pony.ai's autonomous ride-hailing services to global audience. Up to date, Pony.ai has logged over 45 million kilometers in autonomous driving tests worldwide, including more than 8 million kilometers in fully driverless operations.
