SDXL-Lora Lightning is a lightning-fast text-to-image generation lora. It can generate high-quality 1024px images in a few steps. For more information, please refer to our research paper: SDXL-Lightning: Progressive Adversarial Diffusion Distillation . We open-source the model as part of the research.

This lora are distilled from stabilityai/stable-diffusion-xl-base-1.0 . This repository contains checkpoints for 1-step, 2-step, 4-step, and 8-step distilled models. The generation quality of our 2-step, 4-step, and 8-step model is amazing. Our 1-step model is more experimental.

Demos

• Generate with all configurations, best quality: Demo

Diffusers Usage

Please always use the correct checkpoint for the corresponding inference steps.

Recommend Sampler : Euler or DPMPP SDE (work best with sgm_uniform or karras scheduler )
Steps: 8
CFG:1-2.5