Kandinsky 6.0

Kandinsky 6.0 Video is an open-source AI model that generates 5-second video clips with synchronized audio and lip-sync in one pass. Type a prompt, or animate a still image, and it creates the video and the soundtrack together. Two versions: Lite (3B) and Pro (29B), with built-in upscaling to Full-HD. Code, model weights, and diffusers integration are released under the MIT license. Try it on the Hugging Face demo, or run it locally with a powerful GPU.