Alibaba Qwen Open-Sources Qwen-Drive-1.0-4B for Autonomous Driving Perception and Planning

Alibaba's Qwen team has announced the launch of Qwen-Drive-1.0-4B, a vision-language model designed for autonomous driving. Developed in collaboration with Huazhong University of Science and Technology, the model retains the core structure of Qwen3.5-4B while adding new modules for scene understanding and motion planning. The new model allows for free-form inquiries about driving scenes and includes an external BEV perception head that performs 3D object detection and semantic occupancy prediction. The motion planning model is designed to condition on VLM features and generate future trajectories through flow matching.