Alibaba Unveils AI Brains Designed to Power Next Generation of Robots
The suite includes three models and a robot agent framework, with one demo reaching a 196-millisecond inference latency.
- Alibaba Cloud has launched Qwen-Robot Suite, a family of embodied AI models developed by Tongyi Lab that link large language models with real-world robotic actions, currently undergoing pilot testing with enterprise clients.
- The suite comprises specialized models including Qwen-RobotNav for navigation, Qwen-RobotManip for physical interaction, and Qwen-RobotWorld, which predicts environmental changes to enable real-world robotic decision-making.
- Qwen-RobotManip recently achieved the highest score in the RoboChallenge generalist category with a task success rate of 45 percent. In one demonstration, Qwen-RobotNav guided a Unitree Go2 quadruped through an unfamiliar apartment using spoken instructions.
- Alibaba unveiled Qwen-RobotClaw, a framework enabling models to use the suite as physical-world tools, and open-sourced Chat2Robot, a browser-based platform for testing embodied AI interactions.
- Global competition in physical AI accelerates as firms like Google DeepMind and Nvidia expand robotics ecosystems, while start-ups including Figure, Skild, and Physical Intelligence develop general-purpose robotic intelligence.
13 Articles
13 Articles
Alibaba unveils AI brains designed to power next generation of robots
Chinese firm Alibaba has launched its first embodied AI model family, which links large language models with real-world robotic actions. The Qwen-Robot suite was developed by Alibaba’s Tongyi Lab and is undergoing pilot testing with selected Alibaba Cloud enterprise clients. The suite comprises three models focused on navigation, manipulation, and world modeling for robots operating in physical environments. Alibaba said the models enable machin…
Alibaba unveils Qwen-Robot series with three foundation models for embodied AI · TechNode
The Qwen team on Tuesday released a robotics suite featuring three foundation models: Qwen-RobotNav, Qwen-RobotManip, and Qwen-RobotWorld. These three models align language with different types of physical actions. Qwen-RobotNav extends vision-language capabilities into mobile robotics through controllable observation encoding and tool-based interfaces. The model unifies four key tasks within a single framework: instruction following, goal-direc…
[Digital Daily Reporter Gu Ah-hyun] Alibaba Group unveiled the 'Qwen-Robot Suite' on the 17th, a bundle of AI models that endow robots with linguistic, visual, and spatial reasoning capabilities. The Qwen-Robot Suite consists of three models: ▲ Qwen-RobotManip, a Vision-Language-Action (VLA) model specialized for manipulation; ▲ Qwen-RobotNav, a Vision-Language Navigation (VLN) model specialized for mobile navigation; and ▲ Qwen-RobotWorld, a v…
t3nMagazin
HeiseWith the Qwen Robot Suite, the Chinese tech company Alibaba has introduced three AI models specifically for robots, which are designed to move better in the physical world and perform various tasks there. read more on t3n.de
Alibaba unveiled the Qwen-Robot Suite, a set of AI models for robots and physical environments: Qwen-RobotNav for navigation, Qwen-RobotManip for object manipulation, and Qwen-RobotWorld for scene prediction. The team described the project as "a complete stack for embodied artificial intelligence..."
Coverage Details
Bias Distribution
- 67% of the sources are Center
Factuality
To view factuality data please Upgrade to Premium
