On December 28, the acceptance outcomes for the papers submitted to the AAAI 2026 conference were made public. It was revealed that a paper co-authored by XPENG Motors and the National Key Laboratory of Multimedia Information Processing, School of Computer Science, Peking University, had been successfully chosen. This paper presents an efficient visual token pruning framework, FastDriveVLA, tailored specifically for end-to-end autonomous driving Vision-Language-Action (VLA) models. It offers a groundbreaking approach to efficient visual token pruning within the realm of autonomous driving VLA models. XPENG Motors has expressed its commitment to continuing its pursuit of Level 4 autonomous driving capabilities in the future. The company plans to ramp up its investments in the AI large model sector and drive the integration of physical AI large models into automotive applications.