<h1 id="heading-211">ColorfulXL-Lightning</h1><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/3b099dc6-51f8-4e16-a872-39459b34d10d/width=525/3b099dc6-51f8-4e16-a872-39459b34d10d.jpeg" /><h2 id="heading-212">Model Details</h2>ColorfulXL merged with lightning loras (2,8 steps) from bytedance, for fast inference (3-6 steps).Diffusers version: <a target="_blank" rel="ugc" href="https://huggingface.co/recoilme/ColorfulXL-Lightning">https://huggingface.co/recoilme/ColorfulXL-Lightning</a>High range of resolutions supported (576 - 1280), 576*832 example:<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/71a3b50f-4536-4f6f-adb5-9deeab69d850/width=525/71a3b50f-4536-4f6f-adb5-9deeab69d850.jpeg" />Also try Huge resolutions (with Kohya HRFix Integrated, <a target="_blank" rel="ugc" href="https://github.com/lllyasviel/stable-diffusion-webui-forge">SD Forge</a>)<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/c092f946-309b-49d9-8784-3330359a3641/width=525/c092f946-309b-49d9-8784-3330359a3641.jpeg" />Simple inference in 100 line of code <a target="_blank" rel="ugc" href="https://github.com/recoilme/100lineSDXL">https://github.com/recoilme/100lineSDXL</a>Discord (with free generation bot): <a target="_blank" rel="ugc" href="https://discord.com/invite/gsvhQEfKQ5">https://discord.com/invite/gsvhQEfKQ5</a>WIP: <a target="_blank" rel="ugc" href="https://runliq.com/">https://runliq.com/</a><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/73f1d966-d6b7-4d37-9b86-37c1b4be5a15/width=525/73f1d966-d6b7-4d37-9b86-37c1b4be5a15.jpeg" />Due to training LORA's on the base version of SDXL, there are problems with hands and faces, sorry for that:<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/8f20ccf0-5662-4991-aa2a-b97c453165a1/width=525/8f20ccf0-5662-4991-aa2a-b97c453165a1.jpeg" />But you may try ADetailer/Codeformer/HiResFix and so on<h2 id="heading-213">Model Details</h2>Developed by: <a target="_blank" rel="ugc" href="https://aiartlab.org/">https://aiartlab.org/</a>Model type: Diffusion-based text-to-image generative modelModel Description: This model is a fine-tuned model based on colorfulxl <a target="_blank" rel="ugc" href="https://civarchive.com/models/185258/colorfulxl">https://civarchive.com/models/185258/colorfulxl</a>License: This model is not permitted to be used behind API services. Please contact <a target="_blank" rel="ugc" href="mailto:vadim-kulibaba@yandex.ru">vadim-kulibaba@yandex.ru</a> for business inquires, commercial licensing, custom models, and consultation.<h2 id="heading-214">Contact</h2>* For questions and comments about the model, please join <a target="_blank" rel="ugc" href="https://aiartlab.org/](https://aiartlab.org/)">https://aiartlab.org</a>* For future announcements / information about AiArtLab AI models, research, and events, please follow Discord: <a target="_blank" rel="ugc" href="https://discord.com/invite/gsvhQEfKQ5">https://discord.com/invite/gsvhQEfKQ5</a>* For business and partnership inquiries, please contact <a target="_blank" rel="ugc" href="https://t.me/recoilme">https://t.me/recoilme</a>

ColorfulXL-Lightning

<ul><li>colorfulxl_v27+lightning 8 steps 0.25 via super mario</li><li>+lightning 4 steps 0.25 via super mario</li><li>+lightning 2 steps 0.25 via super mario</li><li>colorfulxl_lightning 1.0 + 0.25 result via cosine similarity a</li></ul>

Files with SHA256: 14a3074caef85ad64f8626e5e2678f9c5086abed5ff3daf955242bb91d5e6595