Kandinsky Text-to-Image Diffusion Models

1 repo

Implementations and variants of the Kandinsky diffusion model architecture for text-to-image generation, including the core decoder and prior models, along with specialized extensions like inpainting and depth-based control. These repositories demonstrate the practical application of diffusion-based generative AI, using libraries like Hugging Face diffusers and safetensors for model distribution and inference.

Related clusters