The Academy of Artificial Intelligence (AAI) launched RoboBrain Orca, a multimodal representation world model.

The Beijing Academy of Artificial Intelligence (BAAI) has launched the RoboBrain Orca Multimodal Representation World Model. The model learns unconsciously from 125,000 hours of video and consciously from 160 million event annotations to construct a unified latent representation space for the world. In downstream tasks such as text generation, image prediction, and embodied action generation, the model outperforms baseline models of similar scale.