Alibaba has launched RynnBrain, an open source AI model that helps robots and smart devices perform complex tasks in the real world. The model combines spatial understanding with time awareness.
Alibaba’s DAMO Academy introduced the foundation model that enables interaction with the environment. RynnBrain can map objects, predict trajectories, and navigate in complex environments such as kitchens or factory halls. The system is trained on Alibaba’s Qwen3-VL vision language model.
Alibaba claims state-of-the-art results on benchmarks against Google’s Gemini Robotics-ER 1.5 and Nvidia’s Cosmos-Reason2. The model is available on Hugging Face and GitHub in multiple versions, ranging from 2 billion parameters to a more efficient mixture-of-experts variant.
Open-source versus proprietary
Chinese companies have largely focused on open-source AI. Until now, the open-source strategy in physical AI has been dominated mainly by academic institutions, including Stanford University and the University of California, Berkeley.
By making the model widely available, Alibaba wants to encourage developers and researchers worldwide to refine the technology. This approach could undermine Western leadership by enabling faster innovation cycles.
Alibaba also recently invested $140 million in X Square Robot, a manufacturer of humanoid robots that are already being used in schools, hotels, and healthcare institutions. This demonstrates the Chinese company’s serious ambitions in the robotics race.