Google DeepMind releases Genie 3, an interactive world model

PANews reported on August 5th that Google DeepMind announced the release of Genie 3, a general-purpose world model. This model generates diverse interactive virtual environments in real time based on text prompts, supports dynamic world navigation at 24 frames per second at 720p resolution, and maintains environmental consistency for several minutes. Genie 3 not only excels in physical property modeling and geographical and historical scene reconstruction, but also enables the controllable generation of complex world events. As a key step toward artificial general intelligence ( AGI ), Genie 3 provides a broader simulation space for the training and evaluation of AI somatosensory agents. Currently, the model is available as a limited research preview to selected scholars and creators, with the goal of expanding to more testers in the future.