Kandinsky AI is an open-source multimodal generation tool supporting text-to-image, image-to-image and video generation, with inpainting and blending features, multilingual support, high-resolution output and efficient inference.
