Stable Diffusion 3 Medium Text to Image

Stable Diffusion is a type of latent diffusion model that can generate images from text. It was created by a team of researchers and engineers from CompVis, Stability AI, and LAION. Stable Diffusion v2 is a specific version of the model architecture. It utilizes a downsampling-factor 8 autoencoder with an 865M UNet and OpenCLIP ViT-H/14 text encoder for the diffusion model. When using the SD 2-v model, it produces 768x768 px images. It uses the penultimate text embeddings from a CLIP ViT-H/14 text encoder to condition the generation process.