These workflow templates are intended as multi-purpose templates for use on a wide variety of projects. They can be used with any SDXL checkpoint model.They are intended for use by people that are new to SDXL and ComfyUI.<h3 id="heading-442">List of Templates</h3><ul><li>Simple SDXL Template</li><li>Intermediate SDXL Template<ul><li>ControlNet (Zoe depth)</li></ul></li><li>Advanced SDXL Template<ul><li>ControlNet (4 options)</li><li>A and B versions (see below for more details)</li></ul></li></ul>Additional Simple and Intermediate templates are included, with no Styler node, for users who may be having problems installing the Mile High Styler.<h3 id="heading-1012">Prerequisites</h3><ul><li>ComfyUI installation</li><li>at least 8GB VRAM is recommended</li></ul><h3 id="heading-331">Installation</h3><ul><li>download the Comfyroll SDXL Template Workflows</li><li>download the SDXL models<ul><li><a target="_blank" rel="ugc" href="https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0">SDXL Base 1.0</a></li><li><a target="_blank" rel="ugc" href="https://huggingface.co/stabilityai/stable-diffusion-xl-refiner-1.0">SDXL Refiner 1.0</a></li></ul></li><li>download the SDXL VAE encoder<ul><li><a target="_blank" rel="ugc" href="https://huggingface.co/stabilityai/sdxl-vae">SDXL VAE</a></li></ul></li><li>update ComyUI</li><li>install or update the following custom nodes<ul><li><a target="_blank" rel="ugc" href="https://civarchive.com/models/20793/was-node-suite-comfyui">WAS Node Suite</a></li><li><a target="_blank" rel="ugc" href="https://civarchive.com/models/87609">Comfyroll Custom Nodes</a></li><li><a target="_blank" rel="ugc" href="https://civarchive.com/models/111463">Searge SDXL Nodes</a></li><li><a target="_blank" rel="ugc" href="https://civarchive.com/models/119246">SDXL Style Mile (use latest Ali1234Comfy Extravaganza version)</a></li><li><a target="_blank" rel="ugc" href="https://github.com/Fannovel16/comfy_controlnet_preprocessors">ControlNet Preprocessors by Fannovel16</a><ul><li>needed for preprocessors on the Advanced template</li></ul></li></ul></li><li>ensure you have at least one upscale model installed</li><li>it is recommended to use <a target="_blank" rel="ugc" href="https://civarchive.com/models/71980">ComfyUI Manager</a> for installing and updating custom nodes, for downloading upscale models, and for updating ComfyUI</li><li>the MileHighStyler node is only currently only available via CivitAI</li><li>you may get errors if you have old versions of custom nodes or if ComfyUI is on an old version</li><li>if you have problems installing the preprocessors you can still use the template by removing the preprocessor nodes or replacing with other preprocessor nodes</li></ul>Important: Please run updates on the custom nodes before loading new template versions. This will ensure the template picks up new node versions and not the older versions.Optional downloads (recommended)<ul><li>LoRAs<ul><li><a target="_blank" rel="ugc" href="https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0/tree/main">sd_xl_offset_example-lora_1.0</a></li><li><a target="_blank" rel="ugc" href="https://civarchive.com/models/117060/photorealistic-slider-sdxl-lora">sdxl_photorealistic_slider_v1-0</a></li><li><a target="_blank" rel="ugc" href="https://civarchive.com/models/116480/sdxl10neg4allxlboth-positive-high-qualitydetails-and-negative-worse-qualitybad-hand-in-one-lora">neg4all_bdsqlsz_xl_V6</a></li></ul></li><li>ControlNet<ul><li><a target="_blank" rel="ugc" href="https://huggingface.co/diffusers/controlnet-canny-sdxl-1.0">SDXL 1.0 ControlNet canny</a><ul><li>download diffusion_pytorch_model.fp16.safetensors</li><li>i suggest renaming to canny-xl1.0.safetensors or something similar</li></ul></li><li><a target="_blank" rel="ugc" href="https://huggingface.co/SargeZT/controlnet-sd-xl-1.0-depth-16bit-zoe">SDXL 1.0 ControlNet zoe depth</a><ul><li>download depth-zoe-xl-v1.0-controlnet.safetensors</li></ul></li><li><a target="_blank" rel="ugc" href="https://huggingface.co/thibaud/controlnet-openpose-sdxl-1.0">SDXL 1.0 ControlNet open pose</a><ul><li>download OpenPoseXL2.safetensors</li></ul></li><li><a target="_blank" rel="ugc" href="https://huggingface.co/SargeZT/controlnet-sd-xl-1.0-softedge-dexined">SDXL 1.0 ControlNet softedge-dexined</a><ul><li>download controlnet-sd-xl-1.0-softedge-dexined.safetensors</li></ul></li></ul></li><li>upscale models<ul><li><a target="_blank" rel="ugc" href="https://mega.nz/folder/qZRBmaIY#nIG8KyWFcGNTuMX_XNbJ_g">4x-UltraSharp</a></li><li><a target="_blank" rel="ugc" href="https://huggingface.co/gemasai/4x_NMKD-Siax_200k/tree/main">4x_NMKD-Siax_200k</a></li><li><a target="_blank" rel="ugc" href="https://github.com/xinntao/Real-ESRGAN/releases/download/v0.2.1/RealESRGAN_x2plus.pth">RealESRGAN_x2plus</a></li></ul></li></ul><h3 id="heading-253">Troubleshooting</h3>Please see the new CivitAI article<a target="_blank" rel="ugc" href="https://civarchive.com/articles/1745">Troubleshooting Guide</a><h3 id="heading-332">Installation and Setup</h3>Please see our new CivitAI article<a target="_blank" rel="ugc" href="https://civarchive.com/articles/1901">Installation and Setup Guide</a><h3 id="heading-20">On first use</h3><ul><li>select the XL models and VAE (do not use SD 1.5 models)</li><li>select an upscale model</li><li>add a default image in each of the Load Image nodes (purple nodes)</li><li>add a default image batch in the Load Image Batch node (on Intermediate and Advanced templates)<ul><li>e.g. E:\Comfy Projects\default batch</li><li>it should contain one png image, e.g. image.png</li></ul></li><li>add default LoRAs or set each LoRA to Off and None (on Intermediate and Advanced templates)</li><li>add an XL ControlNet model (on Intermediate and Advanced templates)</li><li>set the batch-size parameter in XL Aspect Ratio (e.g. 1, 2, 4)</li><li>set the filename_prefix in Save Image to your preferred sub-folder</li><li>set control_after_generate in the sampler to randomize</li><li>do a test run</li><li>save a copy to use as your template</li></ul>You can use any SDXL checkpoint model for the Base and Refiner models. Please don't use SD 1.5 models unless you are an advanced user.<h3 id="heading-1048">Things to try (for beginners)</h3><ul><li>try different XL models in the Base model</li><li>try -1 or -2 in CLIP Set Last Layer</li><li>try different styles on the prompt</li><li>try different sampling methods and schedulers in the Sampler</li><li>try different numbers of steps in the Sampler (e.g. 20, 30, 40)</li><li>try different cfg values in the Sampler</li><li>try different base_ratios in the Sampler (e.g. 075, 0.8, 1.0)</li><li>try using Img2Img</li></ul><h3 id="heading-952">A and B Template Versions</h3>A-templates<ul><li>these templates are the easiest to use and are recommended for new users of SDXL and ComfyUI</li><li>they are also recommended for users coming from Auto1111</li><li>the templates produce good results quite easily</li><li>they will also be more stable with changes deployed less often</li></ul>B-templates<ul><li>these templates are 'open beta' WIP templates and will change more often as we try out new ideas</li><li>they are a bit more advanced and so are not recommended for beginners</li><li>they include new SDXL nodes that are being tested out before being deployed to the A-templates</li><li>the default presets are preset 1 and preset A</li></ul><h3 id="heading-667">Simple SDXL Template Features</h3><ul><li>Txt2Img or Img2Img</li><li>batch size on Txt2Img and Img2Img</li><li>SDXL apect ratio selection</li><li>base and refiner models</li><li>separate prompts for positive and negative styles</li><li>SDXL mix sampler</li><li>Hires Fix</li></ul><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/0a137cff-06be-4c77-8113-aa590e6607d8/width=525/0a137cff-06be-4c77-8113-aa590e6607d8.jpeg" /><h3 id="heading-667">Intermediate SDXL Template Features</h3><ul><li>Txt2Img or Img2Img</li><li>batch size on Txt2Img and Img2Img</li><li>SDXL apect ratio selection</li><li>base and refiner models</li><li>separate prompts for potive and negative styles</li><li>SDXL mix sampler</li><li>Hires Fix</li><li>ControlNet zoe depth</li></ul><ul><li>6 LoRA slots (can be toggled On/Off)</li></ul><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/0b069ce4-89fc-47da-b267-d9d7bcdd139c/width=525/0b069ce4-89fc-47da-b267-d9d7bcdd139c.jpeg" /><h3 id="heading-667">Advanced SDXL Template Features</h3><ul><li>Txt2Img or Img2Img</li><li>Img2Img batch</li><li>batch size on Txt2Img and Img2Img</li><li>image padding on Img2Img</li><li>SDXL apect ratio selection</li><li>base and refiner models</li><li>separate prompts for positive and negative styles</li><li>SDXL mix sampler</li><li>Hires Fix</li><li>4 ControlNet options<ul><li>canny</li><li>open pose</li><li>zoe depth</li><li>softedge dexined</li></ul></li></ul><ul><li>6 LoRA slots (can be toggled On/Off)</li><li>post processing options</li></ul>Template A<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/257a5043-a7fc-436f-b4c0-85d70b7a4e0d/width=525/257a5043-a7fc-436f-b4c0-85d70b7a4e0d.jpeg" />Template B<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/3cd3640a-02c2-4ea2-aaa2-e1eaa30eddb9/width=525/3cd3640a-02c2-4ea2-aaa2-e1eaa30eddb9.jpeg" /><h3 id="heading-1051">Tips</h3><ul><li>each optional component can be bypassed using logic switches (red nodes)</li><li>the default setting on all switches is Off (1)</li><li>it is not recommended to use SD1.5 models in the SDXL templates</li><li>some models don't need a refiner so you can set the base_ratio in the sampler to 1 for these</li><li>set Preview method: Auto in ComfyUI Manager to see previews on the samplers</li></ul><h3 id="heading-252">Resources</h3><a target="_blank" rel="ugc" href="https://upscale.wiki/wiki/Model_Database">https://upscale.wiki/wiki/Model_Database</a><a target="_blank" rel="ugc" href="https://github.com/RockOfFire/ComfyUI_Comfyroll_CustomNodes">https://github.com/RockOfFire/ComfyUI_Comfyroll_CustomNodes</a><a target="_blank" rel="ugc" href="https://github.com/SeargeDP/SeargeSDXL">https://github.com/SeargeDP/SeargeSDXL</a><h3 id="heading-2">Change History</h3><ul><li>1.1<ul><li>Advanced Template added<ul><li>A and B versions</li></ul></li><li>new nodes<ul><li>CR Aspect Ratio SDXL</li><li>CR SDXL Prompt Mixer</li><li>CR SDXL Style Text</li><li>CR SDXL Base Prompt Encoder</li><li>CR Img2Img Process Switch</li><li>CR Hires Fix Process Switch</li><li>CR LoRA Stack</li><li>CR Apply LoRA Stack</li><li>CR Latent Batch Size</li><li>CR Halftone Grid</li></ul></li></ul></li></ul><ul><li>2.0<ul><li>Fix for missing Seed node</li><li>new nodes<ul><li>CR Seed</li><li>CR Batch Process Switch</li><li>CR SDXL Prompt Mix Presets</li></ul></li><li>CR SDXL Prompt Mix Presets replaces CR SDXL Prompt Mixer in Advanced Template B</li></ul></li><li>2.1<ul><li>Updated SDXL sampler</li><li>Updated Mile High Styler</li><li>new nodes<ul><li>CR Upscale Image</li></ul></li></ul></li></ul>Advanced stuff starts here - Ignore if you are a beginnerThese are used on SDXL Advanced SDXL Template B only<h3 id="heading-253">SDXL Prompt Presets</h3><ul><li>the prompt presets influence the conditioning applied in the sampler</li><li>there are currently 5 presets</li><li>it is planned to add more presets in future versions</li><li>the detailed preset mappings are provided in the Legacy Mappings section of the article: <a target="_blank" rel="ugc" href="https://civarchive.com/articles/1835">https://civarchive.com/articles/1835</a></li></ul><h3 id="heading-253">SDXL Conditioning Presets</h3><ul><li>the conditioning presets also influence the conditioning applied in the sampler</li><li>there are currently 3 experimental presets</li><li>the default is preset A</li></ul><h3 id="heading-1071">Credits</h3>The prompts used for testing the templates were borrowed from pictures shared on the awesome AI Revolution discord sever.<a target="_blank" rel="ugc" href="https://discord.gg/6PZXWP739k">https://discord.gg/6PZXWP739k</a>There is a thriving ComfyUI community on AI Revolution.The templates owe a lot to the great work done by Searge on developing new SDXL nodes and advanced workflows.

SDXL Workflow Templates for ComfyUI with ControlNet

Files with SHA256: b164d39cb8df8b351b17d10c079e449750d6f052bcefad21350fb1314b0f11d4