This surprisingly versatile SDXL LoRA is trained on a small dataset containing images of the remains of a huge fire at Philips Lighting Application Centre (LAC) Eindhoven, May 14th, 2023. The footage was taken two months later with a Canon G7X mkII point and shoot camera and is <a target="_blank" rel="ugc" href="https://huggingface.co/datasets/leonvanbokhorst/fire-havoc-philips-lac-eindhoven">available on Huggingface</a>. <h2 id="heading-663">Usage</h2>Just drop one of the images into ComfyUI 🥳 or...<ul><li>LoraLoader after the SDXL base model (strength usually at 1.0)</li><li>Both samplers DPM++ SDE</li><li>80 for the base, CFG 10</li><li>20 for the refiner, CFG 6</li><li>Standard SDXL resolutions</li></ul>Many thanks to <a target="_blank" rel="ugc" href="http://Stability.ai">Stability.ai</a> for SDXL, the wonders of <a target="_blank" rel="ugc" href="https://github.com/comfyanonymous/ComfyUI">ComfyUI</a>, and <a target="_blank" rel="ugc" href="https://huggingface.co/TheLastBen">TheLastBen</a> for providing such a comprehensive <a target="_blank" rel="ugc" href="https://github.com/TheLastBen/fast-stable-diffusion">SDXL LoRA training notebook</a> 🙏

Files with SHA256: 17d13301bf016810d26c2737bf96f24cbb5e1eb9ccafc306ed717dfd797af1e4