Home
Alibaba Debuts Qwen-Robot Series with Three Embodied Large Models to Solve Heterogeneous Robot Adaptation

On June 16, Alibaba unveiled the Qwen-Robot series of embodied intelligence models, comprising three core components: the VLA manipulation model Qwen-RobotManip, the VLN navigation model Qwen-RobotNav, and the world model Qwen-RobotWorld. This strategic release reflects a deeper commitment from major tech players to embodied intelligence foundation models, enabling seamless coordination across robot control, navigation, and physical reasoning.
To overcome the industry pain point where traditional VLA models struggle with hardware or scenario changes, Qwen-RobotManip introduces an 80-dimensional unified action representation that defines a universal "body language" for various hardware types, allowing automatic adaptation across devices with just a few feedback steps. The VLN model Qwen-RobotNav, designed for navigation and task execution, builds on Qwen-VL and, for the first time, consolidates five task families—including language-guided navigation, target search, and autonomous driving—into a single framework, removing the need for model switching in complex tasks.
As the cognitive core, Qwen-RobotWorld grants the system the ability to reason about the physical world, predicting and simulating subsequent actions and states. Embodied intelligence is currently transitioning from single-scenario to generalized applications. With all three models released together, Alibaba is poised to accelerate the practical deployment of heterogeneous robots by decoupling the technical architecture and integrating multimodal capabilities.
Related article
Onsemi to Acquire Synaptics for $7 Billion in Stock, Strengthening Position in Edge AI Chip Market
On February 25, leading semiconductor manufacturer onsemi announced its intent to acquire edge AI chip specialist Synaptics in an all-stock deal valued at roughly $7 billion (approximately 47.66 billion yuan). The transaction is projected to close by
Robot Unicorn Yushu Technology Advances to Sci-Tech Innovation Board; 52,000 Yuan upon Signing, Estimated Issue Price Exceeds 100 Yuan
On August 5, China’s leading humanoid robotics firm, Autonavi Technology, initiated its preliminary IPO inquiry on the Science and Technology Innovation Board (STAR Market) between 9:30 and 15:00. The sponsor and lead underwriter simultaneously engag
ChatGPT Introduces Chat Lock Feature for Enhanced Privacy
On Tuesday, OpenAI revealed a new feature allowing users to lock ChatGPT conversations, ensuring sensitive discussions remain private when others might access their screen or device. Users can protect these locked chats using a four-digit PIN, or lev
Related Special Topic Recommendations
Comments (0)
0/500

On June 16, Alibaba unveiled the Qwen-Robot series of embodied intelligence models, comprising three core components: the VLA manipulation model Qwen-RobotManip, the VLN navigation model Qwen-RobotNav, and the world model Qwen-RobotWorld. This strategic release reflects a deeper commitment from major tech players to embodied intelligence foundation models, enabling seamless coordination across robot control, navigation, and physical reasoning.
To overcome the industry pain point where traditional VLA models struggle with hardware or scenario changes, Qwen-RobotManip introduces an 80-dimensional unified action representation that defines a universal "body language" for various hardware types, allowing automatic adaptation across devices with just a few feedback steps. The VLN model Qwen-RobotNav, designed for navigation and task execution, builds on Qwen-VL and, for the first time, consolidates five task families—including language-guided navigation, target search, and autonomous driving—into a single framework, removing the need for model switching in complex tasks.
As the cognitive core, Qwen-RobotWorld grants the system the ability to reason about the physical world, predicting and simulating subsequent actions and states. Embodied intelligence is currently transitioning from single-scenario to generalized applications. With all three models released together, Alibaba is poised to accelerate the practical deployment of heterogeneous robots by decoupling the technical architecture and integrating multimodal capabilities.
Onsemi to Acquire Synaptics for $7 Billion in Stock, Strengthening Position in Edge AI Chip Market
On February 25, leading semiconductor manufacturer onsemi announced its intent to acquire edge AI chip specialist Synaptics in an all-stock deal valued at roughly $7 billion (approximately 47.66 billion yuan). The transaction is projected to close by
Robot Unicorn Yushu Technology Advances to Sci-Tech Innovation Board; 52,000 Yuan upon Signing, Estimated Issue Price Exceeds 100 Yuan
On August 5, China’s leading humanoid robotics firm, Autonavi Technology, initiated its preliminary IPO inquiry on the Science and Technology Innovation Board (STAR Market) between 9:30 and 15:00. The sponsor and lead underwriter simultaneously engag
ChatGPT Introduces Chat Lock Feature for Enhanced Privacy
On Tuesday, OpenAI revealed a new feature allowing users to lock ChatGPT conversations, ensuring sensitive discussions remain private when others might access their screen or device. Users can protect these locked chats using a four-digit PIN, or lev











