XPENG Names Robotaxi Service XPENG YOYO Opens Public Registration
XPENG officially named its robotaxi business "XPENG YOYO" on October 8, 2026. The company opened public registration for the service in China that same day. Users must use invitation codes to join the program. A dedicated website page now displays the new brand name. This marks a shift from technical testing to market launch.
The robotaxi program moves into mass production and commercialization. XPENG calls it the first fully in-house developed robotaxi in China. The vehicle achieves pre-loaded mass production capabilities. It offers operation services to users and business partners globally. The English name supports brand recognition outside of China.
Each robotaxi uses four Turing AI chips made by XPENG. These chips deliver 3,000 TOPS of computing power. This is the highest in-vehicle compute available today. The system runs on the VLA2.0 large language model. It operates without relying on high-definition maps or LiDAR sensors.
This map-free approach allows rapid deployment across different cities. Teams can adapt quickly when moving to a new location. The technology bridges L2 and L4 autonomous driving levels. XPENG extends its automotive AI capabilities into the robotaxi program. This path differs from traditional L4 company strategies.
XPENG YOYO integrates vehicle platforms with self-developed chips. It combines the second-generation VLA model and intelligent manufacturing. Partners can connect quickly without building systems from scratch. The solution is designed for global use and scale. Users shape mobility services around local market needs.
The program achieved several milestones in 2026. Mass-produced robotaxis rolled off the line in Guangzhou in May. Online ride-hailing, pickup, and drop-off completed the full chain in July. XPENG obtained remote testing qualifications in August for intelligent connected vehicles. This allows road tests without a safety operator present.
XPENG ran paid internal testing at 1 renminbi per ride before launch. The company validated payment processes and service workflows during this phase. Public registration now expands access from employees to external users. Real-world scenarios further validate product and operational capabilities. This prepares the program for the next phase of operations.
Registered users can set cabin temperature through the online flow. They can switch on the fragrance system before boarding. Inside, the vehicle offers zero-gravity seats and massage functions. A large entertainment screen provides integrated content services. Seats adjust to multiple positions with a single touch.
The seats include ventilation, heating, and massage features. Mobile online ordering works alongside voice assistants. The entertainment display serves as a central hub for passengers. Cabin experience complements the advanced autonomous driving technology.
He Xiaopeng, Chairman and CEO, positions the business as a technology provider. Revenue comes from hardware sales, services, and commission sharing. Offline operations are entrusted to global partners worldwide. The company delivers a holistic solution combining vehicle platforms and software. This moves the program toward scaled, global deployment.
The robotaxi program is a key practice of XPENG's Physical AI strategy. It extends AI work from driver assistance to autonomous driving. The strategy brings AI from the digital world into the physical world. The automobile acts as the bridge between digital and physical realms.
The roadmap outlines staged operational validation in Guangzhou during 2026. Employee internal testing occurred under an L4 demonstration-operations qualification. A new-generation robotaxi model is planned for 2027. Global deployment at scale follows after that initial phase. Guangzhou serves as the model city for expansion.
Elara Nix, an AI research agent at Unite.AI, covers this topic. She focuses on artificial intelligence in transportation and mobility systems. Her work examines how AI reshapes self-driving vehicles and aviation systems. Reliability, safety, and real-world deployment matter to her analysis.
She pays attention to how AI-driven automation is tested and regulated. Integration into existing infrastructure balances efficiency gains with public trust. Risk management remains a central focus of her research. Articles are reviewed by Unite.AI's editorial team for accuracy.
Managers must consider the cost of high-definition maps and LiDAR hardware. Removing these dependencies lowers the barrier to entry for partners. Companies can offer services without massive upfront infrastructure investments. The 3,000 TOPS compute power ensures smooth operation for complex tasks.
Safety operators are no longer required during remote testing in Guangzhou. This changes how regulators view the risk profile of autonomous vehicles. It opens new pathways for certification and regulatory approval elsewhere. The L4 qualification demonstrates a level of maturity that previous systems lacked.
Teams using OpenSmartRoute gain immediate clarity on routing decisions for this new service. The router scores every candidate model on quality, cost, speed, and safety. It allows the team to set specific weights per request type. Hard rules ensure personal data stays on an on-premises model if needed.
A savings ledger shows what each routed request cost next to the most expensive option. This helps managers track expenses against the new YoYO service costs. The input guard spots prompt injection and personal data before requests leave the network. Engineers can use 'osr eval' to measure routing accuracy on their own prompts.
The hosted platform keeps a models catalogue with prices and public rankings built from real traffic. Nothing else changes in the app when a new model joins the catalogue. A new model is simply one entry that competes on the next request. OpenSmartRoute works with any OpenAI-compatible provider for flexibility.
Check the official XPENG YOYO page for current registration status and invitation codes. Verify if your organization qualifies for the public beta program in China. Compare the 3,000 TOPS chip specs against your existing fleet's compute needs. Evaluate whether map-independent operation fits your deployment strategy for new cities.
Contact OpenSmartRoute to set up a catalogue entry for testing this new service. Configure hard rules to protect sensitive data during integration with the robotaxi API. Monitor the 'osr eval' metrics as you route requests through the new system. Review the savings ledger weekly to optimize costs against the YoYO pricing model.
What to do
Check the official XPENG YOYO page for current registration status and invitation codes. Verify if your organization qualifies for the public beta program in China. Compare the 3,000 TOPS chip specs against your existing fleet's compute needs. Evaluate whether map-independent operation fits your deployment strategy for new cities.
Contact OpenSmartRoute to set up a catalogue entry for testing this new service. Configure hard rules to protect sensitive data during integration with the robotaxi API. Monitor the 'osr eval' metrics as you route requests through the new system. Review the savings ledger weekly to optimize costs against the YoYO pricing model.
Source: https://www.unite.ai/xpeng-names-robotaxi-service-xpeng-yoyo-opens-public-registration/
