World Labs Unveils World Model Atlas, Enabling Multimodal Generation of Text, Images, Audio, and Videos
7 hour ago / Read about 0 minute
Author:小编   

AI enterprise World Labs, established by Li Feifei, has introduced Atlas—the globe's inaugural multimodal world model. This model facilitates pixel-level camera manipulation, has the ability to produce 1440p videos extending up to 1 minute in duration, and excels in spatial reconstruction as well as spatiotemporal simulation.