Alibaba's Qwen Team Launches Qwen-Robot Suite for Robot Automation
The suite marks a push by Alibaba into physical-world AI, offering specialized models for navigation, manipulation, and world understanding that could accelerate robot automation across industries.
Reporting from 1 sources: GIGAZINE.
Alibaba's Qwen AI research team announced Qwen-Robot Suite, a set of three AI models for robot development: Qwen-RobotNav for navigation and autonomous driving, Qwen-RobotManip for operating manipulators, and Qwen-RobotWorld for world recognition. The suite is designed to enable robots to follow natural language instructions and perform complex tasks.
Alibaba's Qwen AI research team announced Qwen-Robot Suite on June 16, 2026, a set of three foundation models for robot development. Qwen-RobotNav, based on Qwen3-VL and trained on 15.6 million samples, handles navigation and target tracking. Qwen-RobotManip, built on Qwen3.5-4B VL and trained on over 38,100 hours of data, operates manipulators with varying structures. Qwen-RobotWorld uses a vision-language model to generate manipulator movements from human demonstration videos and produce multi-angle videos from a single instruction. The suite is intended to support physical AI development.
- Qwen-RobotNav: navigation and autonomous driving, based on Qwen3-VL, 15.6 million training samples
- Qwen-RobotManip: manipulator operation, based on Qwen3.5-4B VL, over 38,100 hours of training data
- Qwen-RobotWorld: world recognition, generates movements from human videos and multi-angle videos from single instruction
Synthesized by Yomimono from the 1 cited source below, including Japanese-language reporting where cited, then editorially reviewed before publishing.