Toyota bZ4X Robotaxi Is in Production, but Its Service Footprint Is Unclear
Pony.ai and Toyota moved the Gen-7 bZ4X robotaxi into mass production in Guangzhou in February 2026, but public information still does not establish how many are carrying paying passengers. Electrive’s February 9 production account placed the vehicle at GAC Toyota’s Guangzhou operation and described a plan to build more than 1,000 units during 2026 for gradual introduction in major Chinese cities.
Pony.ai’s wider robotaxi operation has expanded since that factory milestone, although the company continues to publish combined figures rather than a bZ4X-only tally. In its May 26 operating and financial update, Pony.ai reported a fleet exceeding 1,700 vehicles across Toyota, BAIC and GAC models, raised its year-end target to more than 3,500 vehicles, and disclosed first-quarter robotaxi-service revenue of $8.6 million.
Production is not the same as a completed service launch
The factory milestone established that the Toyota-based robotaxi had progressed beyond a small development batch. The vehicle combines the battery-electric bZ4X platform with Pony.ai’s seventh-generation autonomous-driving system, while GAC Toyota provides an established automotive manufacturing base.
That industrial arrangement matters because a commercial fleet requires repeatable assembly, standardized integration, replacement parts and maintenance support. Producing the autonomous configuration through an automaker’s plant offers a different foundation from retrofitting a limited collection of experimental vehicles.
It does not, however, place every finished car into public service. A robotaxi must pass through production, road testing, local authorization and operational deployment, and those stages can proceed at different speeds in each city.
Guangzhou testing narrows the verified status
By April 24, Pony.ai’s Auto China update said the bZ4X had obtained an on-road testing permit in Guangzhou, retained a plan to deploy 1,000 vehicles during 2026, put the Gen-7 driving kit’s bill-of-materials reduction at 70% from the preceding generation, and projected a complete domestic robotaxi cost of RMB230,000 or less by 2027.
The testing permit is meaningful evidence of regulatory progress, but it should not be read as blanket approval for unrestricted commercial operation. Permissions may differ by operating zone, whether a safety operator is present and whether the vehicle may accept paying passengers.
The cost figures also require careful interpretation. The reduction applies to the autonomous-driving kit’s bill of materials, while the projected amount covers both the kit and base vehicle and remains a future company target; neither figure proves that the bZ4X fleet is already profitable.
The wider fleet data do not reveal bZ4X performance
Pony.ai’s combined fleet numbers show that the company is deploying its latest system at growing scale. They do not disclose the Toyota model’s share because the total also includes the BAIC Arcfox Alpha T5 and GAC Aion V.
The same limitation applies to orders and revenue. Company-wide growth may reflect activity across several vehicle types and operating areas, so it cannot show how many rides the bZ4X completed, how intensively each vehicle was used or which markets had opened the model to ordinary customers.
This distinction prevents two opposite errors. The bZ4X is no longer merely a prototype, but the evidence also falls short of demonstrating a fully deployed, nationwide Toyota robotaxi service.
Why the Toyota partnership still matters
The project tests whether autonomous-driving software can be incorporated into a vehicle program supported by a major manufacturer and its joint-venture production network. Toyota Motor China and GAC Toyota contribute the vehicle and manufacturing capabilities, while Pony.ai supplies the autonomous system and robotaxi operating platform.
That division of work addresses one of the less visible barriers to robotaxi expansion: turning a technically capable vehicle into a maintainable fleet asset. Hardware consistency, redundant vehicle systems, service procedures and parts availability become more important as operators move from test fleets toward routine use.
Lower hardware cost could also improve the economics of expansion, but vehicle cost is only part of the calculation. Charging, cleaning, maintenance, remote assistance, insurance and fleet utilization all affect whether a robotaxi service can operate sustainably.
What remains unknown about the bZ4X rollout
The verified advance is a manufacturing milestone followed by public-road testing, not a fully documented commercial rollout. The available disclosures establish factory production in Guangzhou, regulatory progress there and inclusion of Toyota vehicles in Pony.ai’s growing multi-brand fleet.
They do not provide a completed bZ4X production count, a city-by-city inventory of vehicles in paid driverless service, model-specific ride volumes or operating revenue attributable to the Toyota platform. Those omissions make it impossible to measure the program’s commercial scale from aggregate fleet data alone.
The clearest evidence of the next stage would be model-specific deliveries, commercial driverless permits and operating results tied directly to the bZ4X. Until such data are published, the partnership is best understood as a credible move into automotive-scale production whose passenger-service footprint remains only partially visible.
Also read:
Subscribe to our newsletter
Get the latest Web3, AI, and crypto news delivered straight to your inbox.