SDXL-Lightning is a fast text-to-image generation model that can produce high-quality 1024px images in just a few steps. It is distilled from the stabilityai/stable-diffusion-xl-base-1.0 model and provides checkpoints for 1-step, 2-step, 4-step, and 8-step distilled models. The repository contains both full UNet and LoRA checkpoints, with the full UNet models offering the best quality while the LoRA models can be applied to other base models. SDXL-Lightning can be used with the Diffusers library or ComfyUI, and the authors provide detailed instructions for each use case.