The Chinese AI startup said its new all-modal generative model can create up to 15 seconds of 2K audio-visual content and plans to open-source the model weights in coming days.
26d ago
verifying reliability
No specialized terms available for this topic.