Wan 3.0 AI video generator for creating videos from text, images and multimodal references, with native audio in supported workflows.