Volkswagen will be the launch customer for XPeng’s second-generation Vision-Language-Action model, CEO He Xiaopeng said Wednesday, describing the release as a “DeepSeek moment” for intelligent driving. XPeng unveiled the model in November 2025 and defines it as its first mass-produced physical world large model designed to understand and predict real-world scenarios. The company said the system can perform internal reasoning and simulate responses to different driving scenarios, adding that it increases average driver takeover distance on complex roads by 13 times. [TechNode reporting]
Related
