Tongyi Lab Unveils Unified Vision-Language-Action Model for Embodied AI

Alibaba Group's Tongyi Lab has introduced Qwen-VLA, a new unified Vision-Language-Action model designed for general embodied intelligence. This AI model integrates manipulation, navigation, and trajectory prediction into a single framework, allowing it to perform diverse tasks. It can be deployed across over 10 different robot platforms, including humanoids, without requiring architectural modifications.