Stable Diffusion

An open-weights latent text-to-image diffusion model released by Stability AI in 2022. It operates by iteratively denoising a random latent tensor, conditioned on text embeddings produced by a frozen CLIP encoder, until the latent can be decoded by a VAE into a coherent image. Because its weights and code are openly available, Stable Diffusion has become a common base for research and for accessibility-focused image-generation work — for example, fine-tuning it with LoRA to produce Easy Read pictograms or alt-text illustrations. Practitioners should be aware that the base model reflects biases in its web-scraped training data and is not, without adaptation, suitable for producing cognitively accessible visuals.