<h1 id="heading-480">Control every line!</h1><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/4f880da1-f5cd-4a33-ae39-405c88522cf2/width=525/4f880da1-f5cd-4a33-ae39-405c88522cf2.jpeg" />MistoLine: A Versatile and Robust SDXL-ControlNet Model for Adaptable Line Art ConditioningMistoLine is an SDXL-ControlNet model that can adapt to any type of line art input, demonstrating high accuracy and excellent stability. It can generate high-quality images (with a short side greater than 1024px) based on user-provided line art of various types, including hand-drawn sketches, different ControlNet line preprocessors, and model-generated outlines. MistoLine eliminates the need to select different ControlNet models for different line preprocessors, as it exhibits strong generalization capabilities across diverse line art conditions.We developed MistoLine by employing a novel line preprocessing algorithm (Anyline) and retraining the ControlNet model based on the Unet of stabilityai/stable-diffusion-xl-base-1.0, along with innovations in large model training engineering. MistoLine showcases superior performance across different types of line art inputs, surpassing existing ControlNet models in terms of detail restoration, prompt alignment, and stability, particularly in more complex scenarios.MistoLine maintains consistency with the ControlNet architecture released by lllyasviel, as illustrated in the following schematic diagram:<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/f2165ca7-ab08-464a-9550-2f7827631f13/width=525/f2165ca7-ab08-464a-9550-2f7827631f13.jpeg" /><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/ee2dfc74-27e3-4811-8205-2dc065b7e383/width=525/ee2dfc74-27e3-4811-8205-2dc065b7e383.jpeg" />reference：https://github.com/lllyasviel/ControlNet More information about ControlNet can be found in the following references:<a target="_blank" rel="ugc" href="https://github.com/lllyasviel/ControlNet">https://github.com/lllyasviel/ControlNet</a><a target="_blank" rel="ugc" href="https://huggingface.co/docs/diffusers/main/en/api/pipelines/controlnet_sdxl">https://huggingface.co/docs/diffusers/main/en/api/pipelines/controlnet_sdxl</a>The model is compatible with most SDXL models, except for PlaygroundV2.5 and CosXL. It can be used in conjunction with LCM and other ControlNet models. We have open-sourced the corresponding model weight files for non-commercial use by individual users.<h2 id="heading-481">Apply with different line preprocessor</h2><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/a35195ff-8fba-45e9-91d2-53f933f0be90/width=525/a35195ff-8fba-45e9-91d2-53f933f0be90.jpeg" /><h2 id="heading-482">Mistoline compere with other Controlnet</h2><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/bed54609-59cd-487d-94bd-8f205d28dd15/width=525/bed54609-59cd-487d-94bd-8f205d28dd15.jpeg" /><h2 id="heading-483">Application examples</h2><h3 id="heading-484">Sketch rendering</h3>The following case only utilized MistoLine as the controlnet：<h3 id="heading-485"><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/f591507f-9ba0-4578-9d61-9ee70d9a7ce7/width=525/f591507f-9ba0-4578-9d61-9ee70d9a7ce7.jpeg" /></h3><h3 id="heading-486">Model rendering</h3>The following case only utilized Anyline as the preprocessor and MistoLine as the controlnet.<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/33dd0f9c-3e73-42f4-ab46-9a47167f3db6/width=525/33dd0f9c-3e73-42f4-ab46-9a47167f3db6.jpeg" />ComfyUI Recommended Parameters: sampler steps:30CFG:7.0sampler_name:dpmpp_2m_sdescheduler:karrasdenoise:0.93controlnet_strength:1.0stargt_percent:0.0end_percent:0.9Checkpoints• mistoLine_rank256.safetensors : General usage version, for <a target="_blank" rel="ugc" href="https://github.com/comfyanonymous/ComfyUI">ComfyUI</a> and <a target="_blank" rel="ugc" href="https://github.com/AUTOMATIC1111">AUTOMATIC1111</a>-WebUI.• mistoLine_fp16.safetensors : FP16 weights, for <a target="_blank" rel="ugc" href="https://github.com/comfyanonymous/ComfyUI">ComfyUI</a> and <a target="_blank" rel="ugc" href="https://github.com/AUTOMATIC1111">AUTOMATIC1111</a>-WebUI.ComfyUI Usage<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/7d04f6e9-8c50-4e06-b178-8a70fcaef227/width=525/7d04f6e9-8c50-4e06-b178-8a70fcaef227.jpeg" />中国（大陆地区）便捷下载地址：链接：<a target="_blank" rel="ugc" href="https://pan.baidu.com/s/1DbZWmGJ40Uzr3Iz9RNBG_w?pwd=8mzs">https://pan.baidu.com/s/1DbZWmGJ40Uzr3Iz9RNBG_w?pwd=8mzs</a>提取码：8mzsCitation @misc{ title={Adding Conditional Control to Text-to-Image Diffusion Models}, author={Lvmin Zhang, Anyi Rao, Maneesh Agrawala}, year={2023}, eprint={2302.05543}, archivePrefix={arXiv}, primaryClass={<a target="_blank" rel="ugc" href="http://cs.CV">cs.CV</a>}}

Files with SHA256: 6c46c871a49bf9e74114f0e1f87d818426203fd62847e78043e4ce2d40834d53