High likelihood is that I am misunderstanding how I use both in conjunction within comfy. I tried with two checkpoint combinations but got the same results : sd_xl_base_0. In the Comfyui SDXL workflow example, the refiner is an integral part of the generation process. 9. 2. . So, with a little bit of effort it is possible to get ComfyUI up and running alongside your existing Automatic1111 install and to push out some images from the new SDXL model. I've been using SDNEXT for months and have had NO PROBLEM. You don't need refiner model in custom. 0 workflow. make a folder in img2img. 5 renders, but the quality i can get on sdxl 1. silenf • 2 mo. The sample prompt as a test shows a really great result. x, SD2. Do you have ComfyUI manager. Members Online •. 5 checkpoint files? currently gonna try them out on comfyUI. While SDXL offers impressive results, its recommended VRAM (Video Random Access Memory) requirement of 8GB poses a challenge for many. 9 safetensors + LoRA workflow + refiner The text was updated successfully, but these errors were encountered: All reactionsComfyUI got attention recently because the developer works for StabilityAI and was able to be the first to get SDXL running. In the Kohya interface, go to the Utilities tab, Captioning subtab, then click WD14 Captioning subtab. To simplify the workflow set up a base generation and refiner refinement using two Checkpoint Loaders. So I have optimized the ui for SDXL by removing the refiner model. 9 workflow, the one that olivio sarikas video works just fine) just replace the models with 1. Template Features. Closed BitPhinix opened this issue Jul 14, 2023 · 3. 5s/it, but the Refiner goes up to 30s/it. The test was done in ComfyUI with a fairly simple workflow to not overcomplicate things. The workflow should generate images first with the base and then pass them to the refiner for further refinement. . 0 base and refiner models with AUTOMATIC1111's Stable Diffusion WebUI. SDXL 1. It provides a super convenient UI and smart features like saving workflow metadata in the resulting PNG images. @bmc-synth You can use base and/or refiner to further process any kind of image, if you go through img2img (out of latent space) and proper denoising control. 0 Alpha + SD XL Refiner 1. x, SD2. When all you need to use this is the files full of encoded text, it's easy to leak. 0 Base and Refiners models downloaded and saved in the right place, it should work out of the box. — NOTICE: All experimental/temporary nodes are in blue. The refiner is although only good at refining noise from an original image still left in creation, and will give you a blurry result if you try to add. • 3 mo. 0 with the node-based user interface ComfyUI. At that time I was half aware of the first you mentioned. July 4, 2023. 9 fine, but when I try to add in the stable-diffusion-xl-refiner-0. json file which is easily loadable into the ComfyUI environment. 0 base and refiner and two others to upscale to 2048px. See "Refinement Stage" in section 2. 0 Base model used in conjunction with the SDXL 1. ComfyUIを使ってみる勇気 以上です。 「なんか難しそうで怖い…🥶」という方は、まず私の動画を見てComfyUIのイメトレをしてから望むのも良いと思います。I just wrote an article on inpainting with SDXL base model and refiner. . SDXL Lora + Refiner Workflow. 0 with both the base and refiner checkpoints. 启动Comfy UI. 10:05 Starting to compare Automatic1111 Web UI with ComfyUI for SDXL. 17:38 How to use inpainting with SDXL with ComfyUI. The SDXL base model performs significantly better than the previous variants, and the model combined with the refinement module achieves the best overall performance. sd_xl_refiner_0. If you use ComfyUI and the example workflow that is floading around for SDXL, you need to do 2 things to resolve it. workflow custom-nodes stable-diffusion comfyui sdxl Updated Nov 13, 2023; Python;. It makes it really easy if you want to generate an image again with a small tweak, or just check how you generated something. 33. 999 RC August 29, 2023. 0 Base model used in conjunction with the SDXL 1. Step 4: Copy SDXL 0. 第一、风格控制 第二、base模型以及refiner模型如何连接 第三、分区提示词控制 第四、多重采样的分区控制 comfyui节点流程这个东西一通百通，逻辑正确怎么连都可以，所以这个视频我讲得并不仔细. Now in Comfy, from the Img2img workflow, let’s duplicate Load Image and Upscale Image Nodes. There is an SDXL 0. 9 Refiner. The workflow should generate images first with the base and then pass them to the refiner for further. image padding on Img2Img. Create and Run SDXL with SDXL. It makes it really easy if you want to generate an image again with a small tweak, or just check how you generated something. source_folder_path = '/content/ComfyUI/output' # Replace with the actual path to the folder in th e runtime environment destination_folder_path = f '/content/drive/MyDrive/ {output_folder_name} ' # Replace with the desired destination path in you r Google Drive # Create the destination folder in Google Drive if it doesn't existI wonder if it would be possible to train an unconditional refiner that works on RGB images directly instead of latent images. Make the following changes: In the Stable Diffusion checkpoint dropdown, select the refiner sd_xl_refiner_1. Includes LoRA. In any case, just grabbing SDXL. BTW, Automatic1111 and ComfyUI won't give you the same images except you changes some settings on Automatic1111 to match ComfyUI because the seed generation is different as far as I Know. see this workflow for combining SDXL with a SD1. Per the. Place VAEs in the folder ComfyUI/models/vae. json file which is easily loadable into the ComfyUI environment. 0 Base should have at most half the steps that the generation has. Mostly it is corrupted if your non-refiner works fine. The SDXL base checkpoint can be used like any regular checkpoint in ComfyUI. Efficiency Nodes for ComfyUI A collection of ComfyUI custom nodes to help streamline workflows and reduce total node count. To use the refiner model: Navigate to the image-to-image tab within AUTOMATIC1111 or. 4. So if ComfyUI / A1111 sd-webui can't read the image metadata, open the last image in a text editor to read the details. . 0; the highly-anticipated model in its image-generation series! After you all have been tinkering away with randomized sets of models on our Discord bot, since early May, we’ve finally reached our winning crowned-candidate together for the release of SDXL 1. web UI（SD. Reply reply1. 9版本的base model,refiner model. 手順5：画像を生成. ComfyUI provides a powerful yet intuitive way to harness Stable Diffusion through a flowchart interface. Hires fix is just creating an image at a lower resolution, upscaling it and then sending it through img2img. Sytan SDXL ComfyUI: Very nice workflow showing how to connect the base model with the refiner and include an upscaler. Currently, a beta version is out, which you can find info about at AnimateDiff. 0_0. Think of the quality of 1. For example: 896x1152 or 1536x640 are good resolutions. Thanks. 0. With usable demo interfaces for ComfyUI to use the models (see below)! After test, it is also useful on SDXL-1. 5 and 2. Part 4 (this post) - We will install custom nodes and build out workflows. WAS Node Suite. that should stop it being distorted, you can also switch the upscale method to bilinear as that may work a bit better. Working amazing. 0 Base+Refiner比较好的有26. launch as usual and wait for it to install updates. download the SDXL VAE encoder. Do I need to download the remaining files pytorch, vae and unet? also is there an online guide for these leaked files or do they install the same like 2. To use this workflow, you will need to set. • 3 mo. It also lets you specify the start and stop step which makes it possible to use the refiner as intended. Model type: Diffusion-based text-to-image generative model. A hub dedicated to development and upkeep of the Sytan SDXL workflow for ComfyUI he workflow is provided as a . 10. 9) Tutorial | Guide 1- Get the base and refiner from torrent. . 2 comments. 1 for ComfyUI. Table of Content. Developed by: Stability AI. Here's a simple workflow in ComfyUI to do this with basic latent upscaling: The big current advantage of ComfyUI over Automatic1111 is it appears to handle VRAM much better. I just downloaded the base model and the refiner, but when I try to load the model it can take upward of 2 minutes, and rendering a single image can take 30 minutes, and even then the image looks very very weird. 4/1. SDXL_1 (right click and save as) workflow has the SDXL setup with refiner with best settings. The two-model setup that SDXL uses has the base model is good at generating original images from 100% noise, and the refiner is good at adding detail at 0. SDXL places very heavy emphasis at the beginning of the prompt, so put your main keywords. The ONLY issues that I've had with using it was with the. 0 ComfyUI. Detailed install instruction can be found here: Link to. Download and drop the JSON file into ComfyUI. 0 Base Only 多出4%左右 Comfyui工作流：Base onlyBase + RefinerBase + lora + Refiner SD1. 手順3：ComfyUIのワークフローを読み込む. x. 20:43 How to use SDXL refiner as the base model. For those of you who are not familiar with ComfyUI, the workflow (images #3) appears to be: Generate text2image "Picture of a futuristic Shiba Inu", with negative prompt "text, watermark" using SDXL base 0. With Tiled Vae (im using the one that comes with multidiffusion-upscaler extension) on, you should be able to generate 1920x1080, with Base model, both in txt2img and img2img. Here's what I've found: When I pair the SDXL base with my LoRA on ComfyUI, things seem to click and work pretty well. Create and Run Single and Multiple Samplers Workflow, 5. 0. Hello! A lot has changed since I first announced ComfyUI-CoreMLSuite. By becoming a member, you'll instantly unlock access to 67 exclusive posts. But suddenly the SDXL model got leaked, so no more sleep. . Fine-tuned SDXL (or just the SDXL Base) All images are generated just with the SDXL Base model or a fine-tuned SDXL model that requires no Refiner. T2I-Adapter aligns internal knowledge in T2I models with external control signals. I've been tinkering with comfyui for a week and decided to take a break today. The SDXL base model performs significantly better than the previous variants, and the model combined with the refinement module achieves the best overall performance. 0 in ComfyUI I've come across three different methods that seem to be commonly used: Base Model with Latent Noise Mask, Base Model using InPaint VAE Encode and using the UNET "diffusion_pytorch" InPaint specific model from Hugging Face. All the images in this repo contain metadata which means they can be loaded into ComfyUI with the Load button (or dragged onto the window) to get the full workflow that was used to create the image. I found it very helpful. 次にSDXLのモデルとVAEをダウンロードします。 SDXLのモデルは2種類あり、基本のbaseモデルと、画質を向上させるrefinerモデルです。 どちらも単体で画像は生成できますが、基本はbaseモデルで生成した画像をrefinerモデルで仕上げるという流れが一般的なよう. SD-XL 0. png . Traditionally, working with SDXL required the use of two separate ksamplers—one for the base model and another for the refiner model. download the Comfyroll SDXL Template Workflows. Specialized Refiner Model: SDXL introduces a second SD model specialized in handling high-quality, high-resolution data;. Aug 2. After completing 20 steps, the refiner receives the latent space. To simplify the workflow set up a base generation and refiner refinement using two Checkpoint Loaders. x, SDXL and Stable Video Diffusion; Asynchronous Queue systemComfyUI installation. ·. 5 models) to do. 0 refiner model. It will destroy the likeness because the Lora isn’t interfering with the latent space anymore. png","path":"ComfyUI-Experimental. sdxl-0. Restart ComfyUI. 2 Workflow - Simple - Easy to use and simple with Upscaling 4K, just. set COMMANDLINE_ARGS=--medvram --no-half-vae --opt-sdp-attention. Refiner > SDXL base > Refiner > RevAnimated, to do this in Automatic1111 I would need to switch models 4 times for every picture which takes about 30 seconds for each switch. For those of you who are not familiar with ComfyUI, the workflow (images #3) appears to be: Generate text2image "Picture of a futuristic Shiba Inu", with negative prompt "text,. if you find this helpful consider becoming a member on patreon, subscribe to my youtube for Ai applications guides. 5. Join to Unlock. Example script for training a lora for the SDXL refiner #4085. 0; the highly-anticipated model in its image-generation series! After you all have been tinkering away with randomized sets of models on our Discord bot, since early May, we’ve finally reached our winning crowned-candidate together for the release of SDXL 1. In part 1 ( link ), we implemented the simplest SDXL Base workflow and generated our first images. SDXL 1. update ComyUI. 6. 3. SDXL Refiner 1. 0. Images. 1. com Open. ComfyUI seems to work with the stable-diffusion-xl-base-0. If this is. . Direct Download Link Nodes: Efficient Loader &. Then move it to the “ComfyUImodelscontrolnet” folder. ) Sytan SDXL ComfyUI. Updated with 1. Exciting SDXL 1. safetensors files to the ComfyUI file which is present with name ComfyUI_windows_portable file. base and refiner models. r/StableDiffusion. 5 models. SDXL Examples The SDXL base checkpoint can be used like any regular checkpoint in ComfyUI. 0 with ComfyUI. The refiner model works, as the name suggests, a method of refining your images for better quality. ComfyUI also has a mask editor that can be accessed by right clicking an image in the LoadImage node and "Open in MaskEditor". 9 (just search in youtube sdxl 0. And I'm running the dev branch with the latest updates. Got playing with SDXL and wow! It's as good as they stay. You must have sdxl base and sdxl refiner. 5 comfy JSON and import it sd_1-5_to_sdxl_1-0. SDXL Workflow for ComfyBox - The power of SDXL in ComfyUI with better UI that hides the nodes graph Resource | Update I recently discovered ComfyBox, a UI fontend for ComfyUI. I tried comfyUI and it takes about 30s to generate 768*1048 images (i have a RTX2060, 6GB vram). python launch. 5x), but I can't get the refiner to work. I can run SDXL 1024 on comfyui with an 2070/8GB smoother than I could run 1. 0 Refiner model. The solution to that is ComfyUI, which could be viewed as a programming method as much as it is a front end. Upscale model, (needs to be downloaded into ComfyUImodelsupscale_models Recommended one is 4x-UltraSharp, download from here. Here are the configuration settings for the SDXL. ( I am unable to upload the full-sized image. SD+XL workflows are variants that can use previous generations. Installing ControlNet. 9vae Image size: 1344x768px Sampler: DPM++ 2s Ancestral Scheduler: Karras Steps: 70 CFG Scale: 10 Aesthetic Score: 6It would be neat to extend the SDXL dreambooth Lora script with an example of how to train the refiner. Step 2: Install or update ControlNet. 0以降 である必要があります（※もっと言うと後述のrefinerモデルを手軽に使うためにはv1. I mean, it's also possible to use it like that, but the proper intended way to use the refiner is a two-step text-to-img. 3. This uses more steps, has less coherence, and also skips several important factors in-between. 2 more replies. The SDXL Prompt Styler is a versatile custom node within Comfy UI that streamlines the prompt styling process. . It fully supports the latest Stable Diffusion models including SDXL 1. Embeddings/Textual Inversion. md","path":"README. 0 and refiner) I can generate images in 2. Usage Notes SDXL two staged denoising workflow. One of its key features is the ability to replace the {prompt} placeholder in the ‘prompt’ field of these. The lost of details from upscaling is made up later with the finetuner and refiner sampling. 2xxx. x, SDXL, LoRA, and upscaling makes ComfyUI flexible. 9. My comfyui is updated and I have latest versions of all custom nodes. 5 and 2. 0 and upscalers. The initial image in the Load Image node. SDXL_1 (right click and save as) workflow has the SDXL setup with refiner with best settings. Apprehensive_Sky892. This notebook is open with private outputs. 0. 0 seed: 640271075062843 大家好，我是小志Jason。一个探索Latent Space的程序员。今天来深入讲解一下SDXL的工作流，顺便说一下SDXL和过去的SD流程有什么区别 官方在discord上chatbot测试的数据，文生图觉得SDXL 1. that extension really helps. I upscaled it to a resolution of 10240x6144 px for us to examine the results. New comments cannot be posted. Download both from CivitAI and move them to your ComfyUI/Models/Checkpoints folder. To make full use of SDXL, you'll need to load in both models, run the base model starting from an empty latent image, and then run the refiner on the base model's output to improve detail. I’ve created these images using ComfyUI. At that time I was half aware of the first you mentioned. Load成功后你应该看到的是这个界面，你需要重新选择一下你的refiner和base modelI was having very poor performance running SDXL locally in ComfyUI to the point where it was basically unusable. Especially on faces. I was able to find the files online. 5 model (directory: models/checkpoints) Install your loras (directory: models/loras) Restart. 20:43 How to use SDXL refiner as the base model. For instance, if you have a wildcard file called. The SDXL base checkpoint can be used like any regular checkpoint in ComfyUI. 0 Refiner Automatic calculation of the steps required for both the Base and the Refiner models Quick selection of image width and height based on the SDXL training set XY Plot ControlNet with the XL OpenPose model (released by Thibaud Zamora)ComfyUI may take some getting used to, mainly as it is a node-based platform, requiring a certain level of familiarity with diffusion models. 0 Part 5: Scale and Composite Latents with SDXL Part 6: SDXL 1. SDXL Base 1. ComfyUI is a powerful modular graphic interface for Stable Diffusion models that allows you to create complex workflows using nodes. Fine-tuned SDXL (or just the SDXL Base) All images are generated just with the SDXL Base model or a fine-tuned SDXL model that requires no Refiner. It will crash eventually - possibly RAM but doesn't take the VM with it - but as a comparison that one "works". I’m going to discuss…11:29 ComfyUI generated base and refiner images. Text2Image with SDXL 1. 9モデル2つ（BASE, Refiner) 2. If you want a fully latent upscale, make sure the second sampler after your latent upscale is above 0. FWIW latest ComfyUI does launch and renders some images with SDXL on my EC2. ComfyUI with SDXL (Base+Refiner) + ControlNet XL OpenPose + FaceDefiner (2x) ComfyUI is hard. Upcoming features:This is the image I created using ComfyUI, utilizing Dream ShaperXL 1. ⚠️ IMPORTANT: Due to shifts in priorities and a decreased interest in this project from my end, this repository will no longer receive updates or maintenance. The chart above evaluates user preference for SDXL (with and without refinement) over SDXL 0. If the refiner doesn't know the LoRA concept any changes it makes might just degrade the results. I've a 1060 GTX, 6gb vram, 16gb ram. Must be the architecture. 9-usage This repo is a tutorial intended to help beginners use the new released model, stable-diffusion-xl-0. 0. I've been having a blast experimenting with SDXL lately. These are examples demonstrating how to do img2img. SDXL 1. Inpainting. u/EntrypointjipDiscover the Ultimate Workflow with ComfyUI in this hands-on tutorial, where I guide you through integrating custom nodes, refining images with advanced tool. 5 models and I don't get good results with the upscalers either when using SD1. まず大きいのがSDXLの Refiner機能 に対応しました。 以前も紹介しましたが、SDXL では 2段階 での画像生成方法を取り入れています。 まず Baseモデル で構図などの絵の土台を作成し、 Refinerモデル で細部のディテールを上げることでクオリティの高. However, with the new custom node, I've. cd ~/stable-diffusion-webui/. You generate the normal way, then you send the image to imgtoimg and use the sdxl refiner model to enhance it. 9-base Model のほか、SD-XL 0. import json from urllib import request, parse import random # this is the ComfyUI api prompt format. It is a Latent Diffusion Model that uses a pretrained text encoder ( OpenCLIP-ViT/G ). About SDXL 1. 0—a remarkable breakthrough. 追記：2023/09/20 Google Colab の無料枠でComfyuiが使えなくなったため、別のGPUサービスを使ってComfyuiを起動するNotebookを作成しました。 記事の後半で解説していきます。 今回は、 Stable Diffusion Web UI のようにAIイラストを生成できる ComfyUI というツールを使って、簡単に AIイラスト を生成する方法ご. What a move forward for the industry. I have updated the workflow submitted last week, cleaning up a bit the layout and adding many functions I wanted to learn better. 3. 0下载公布,本机部署教学-A1111+comfyui，共用模型，随意切换|SDXL SD1. 2 Workflow - Face - for Base+Refiner+VAE, FaceFix and Upscaling 4K; 1. SDXL-ComfyUI-Colab One click setup comfyUI colab notebook for running SDXL (base+refiner). utils import load_image pipe = StableDiffusionXLImg2ImgPipeline. 5 checkpoint files? currently gonna try them out on comfyUI. Model loaded in 5. ComfyUI is great if you're like a developer because you can just hook up some nodes instead of having to know Python to update A1111. Nodes/graph/flowchart interface to experiment and create complex Stable Diffusion workflows without needing to code anything. I wanted to share my configuration for ComfyUI, since many of us are using our laptops most of the time. +Use SDXL Refiner as Img2Img and feed your pictures. Opening_Pen_880. 0 through an intuitive visual workflow builder. 0 Checkpoint Models beyond the base and refiner stages. You can Load these images in ComfyUI to get the full workflow. Always use the latest version of the workflow json file with the latest version of the custom nodes!Yes it’s normal, don’t use refiner with Lora. Users can drag and drop nodes to design advanced AI art pipelines, and also take advantage of libraries of existing workflows. bat file to the same directory as your ComfyUI installation. Hotshot-XL is a motion module which is used with SDXL that can make amazing animations. Generating 48 in batch sizes of 8 in 512x768 images takes roughly ~3-5min depending on the steps and the sampler. It takes around 18-20 sec for me using Xformers and A111 with a 3070 8GB and 16 GB ram. I had experienced this too, dont know checkpoint is corrupted but it’s actually corrupted Perhaps directly download into checkpoint folderI tried SDXL in A1111, but even after updating the UI, the images take veryyyy long time and don't finish, like they stop at 99% every time. That's the one I'm referring to. If you want to open it. 这一期呢我们来开个新坑，来讲sd的另一种打开方式，也就是这个节点化comfyUI。那熟悉我们频道的老观众都知道，我一直是用webUI去做演示和讲解的. 0 workflow. ComfyUI-CoreMLSuite now supports SDXL, LoRAs and LCM. Part 1: Stable Diffusion SDXL 1. Searge-SDXL: EVOLVED v4. 1 - Tested with SDXL 1. SDXL - The Best Open Source Image Model. 0 Resource | Update civitai. Yes, 8Gb card, ComfyUI workflow loads both SDXL base & refiner models, separate XL VAE, 3 XL LoRAs, plus Face Detailer and its sam model and bbox detector model, and Ultimate SD Upscale with its ESRGAN model and input from the same base SDXL model all work together. at least 8GB VRAM is recommended. SDXL comes with a base and a refiner model so you’ll need to use them both while generating images. I did extensive testing and found that at 13/7, the base does the heavy lifting on the low-frequency information, and the refiner handles the high-frequency information, and neither of them interferes with the other's specialtysdxl_v1. Well, SDXL has a refiner, I'm sure you're asking right about now - how do we get that implemented? Although SDXL works fine without the refiner (as demonstrated above) you really do need to use the refiner model to get the full use. Today, I upgraded my system to 32GB of RAM and noticed that there were peaks close to 20GB of RAM usage, which could cause memory faults and rendering slowdowns in a 16gb system. Tutorial Video : ComfyUI Master Tutorial - Stable Diffusion XL (SDXL) - Install On PC, Google Colab (Free) & RunPod . If you're using ComfyUI you can right click on a Load Image node and select "Open in MaskEditor" to draw an inpanting mask. 0 ComfyUI Workflow With Nodes Use Of SDXL Base & Refiner ModelIn this tutorial, join me as we dive into the fascinating worl. It is highly recommended to use a 2x upscaler in the Refiner stage, as 4x will slow the refiner to a crawl on most systems, for no significant benefit (in my opinion). You can disable this in Notebook settingsComfyUI is a powerful and modular GUI for Stable Diffusion that lets you create advanced workflows using a node/graph interface. json file to ComfyUI window. SDXL-refiner-1. 9 and Stable Diffusion 1. Create a Load Checkpoint node, in that node select the sd_xl_refiner_0. SDXL apect ratio selection. 0 links. Stable Diffusion is a Text to Image model, but this sounds easier than what happens under the hood. Adjust the workflow - Add in the. 0 for ComfyUI, today I want to compare the performance of 4 different open diffusion models in generating photographic content: SDXL 1. Fine-tuned SDXL (or just the SDXL Base) All images are generated just with the SDXL Base model or a fine-tuned SDXL model that requires no Refiner. SDXL VAE. By default, AP Workflow 6. Explain COmfyUI Interface Shortcuts and Ease of Use. batch size on Txt2Img and Img2Img. SDXL Models 1. 5 works with 4GB even on A1111 so you either don't know how to work with ComfyUI or you have not tried it at all. 0 refiner checkpoint; VAE. X etc. x models through the SDXL refiner, for whatever that's worth! Use Loras, TIs, etc, in the style of SDXL, and see what more you can do. For upscaling your images: some workflows don't include them, other workflows require them. Lora. Most UI's req. 3-中文必备插件篇,stable diffusion教学，stable diffusion进阶教程3：comfyui深度体验以及照片转漫画工作流详解，ComfyUI系统性教程来啦！简体中文版整合包+全新升级云部署！预装超多模块组一键启动！All images were created using ComfyUI + SDXL 0. 0. Play around with different Samplers and different amount of base Steps (30, 60, 90, maybe even higher). Voldy still has to implement that properly last I checked. Yup, all images generated in the main ComfyUI frontend have the workflow embedded into the image like that (right now anything that uses the ComfyUI API doesn't have that, though). ComfyUI_00001_. Most ppl use ComfyUI which is supposed to be more optimized than A1111 but for some reason, for me, A1111 is more faster, and I love the external network browser to organize my Loras. Workflow for ComfyUI and SDXL 1. Tutorial Video : ComfyUI Master Tutorial - Stable Diffusion XL (SDXL) - Install On PC, Google Colab. 8s (create model: 0. Source. I've been having a blast experimenting with SDXL lately. Holding shift in addition will move the node by the grid spacing size * 10. Just wait til SDXL-retrained models start arriving. 0! Usage This workflow is meticulously fine tuned to accommodate LORA and Controlnet inputs, and demonstrates interactions with embeddings as well. Once wired up, you can enter your wildcard text. Comfyroll. 23:06 How to see ComfyUI is processing the which part of the. 0 base checkpoint; SDXL 1. webui gradio sd stable-diffusion stablediffusion stable-diffusion-webui sdxl Updated Oct 28 , 2023.