Add models to comfyui 4x_NMKD-Siax_200k. safetensors model installed, you’ll see this prompt: 5 days ago · In ComfyUI, ControlNet models for different versions of drawing models are not compatible with each other. Includes detailed instructions for model installation and parameter adjustments. 0 model file that you downloaded. Perhaps using a detailer model might help? I've invested a lot of time into this with no success. Put model files in your Google Drive. Mar 10, 2025 · #Rename this to extra_model_paths. Aug 29, 2023 · Configuring Models Location for ComfyUI. 1 ComfyUI Workflow. Note: Remember to add your models, VAE, LoRAs etc. yaml and ComfyUI will load it #config for a1111 ui #all you have to do is change the base_path to where yours is installed a111: base_path: path/to/stable-diffusion-webui/ checkpoints: models/Stable-diffusion configs: models/Stable-diffusion vae: models/VAE loras: | models/Lora models/LyCORIS upscale_models Run ComfyUI workflows online and deploy APIs with one click Dec 6, 2024 · Everything here is the same (generally) as the top branch, but it uses the Create Hook Model as LoRA node instead of the Create Hook LoRA node. "I left the name as is, as ComfyUI Manger put the file into a named folder : -)))) You can also subtract models weights and add them like in this example used to create an inpaint model from a non inpaint model with the formula: (inpaint_model - base_model) * 1. to the corresponding Comfy folders, as discussed in ComfyUI manual installation. pth upscaler; 4x-Ultrasharp 2 days ago · A comprehensive beginner's guide to using LoRA in ComfyUI, covering both basic single model usage and advanced multiple LoRA combinations to achieve richer AI art effects. As a verb, to model something means to represent it or provide an example. py --directml Mar 22, 2024 · #Rename this to extra_model_paths. This guide will steer you through the process of installing additional models from the ComfyUI Extra Models repository. For upscaling your images: some workflows don't include them, other workflows require them. Apr 11, 2025 · Below is a guide on installing and using the Stable Diffusion model in ComfyUI. Rename it to. Dec 19, 2023 · Place VAEs in the folder ComfyUI/models/vae. Belittling their efforts will get you banned. In this video we will understand what is a checkpoint. yaml after launching ComfyUI Desktop, under the menu Help -> Open Folder -> Open extra_model_paths. Written by comfyanonymous and other contributors. After installation, you can start using ControlNet models in ComfyUI. 2 Download Models from Hugging Face. To understand better, read the below link talking about the sampler types. Download the model you want to use. safetensors and place the model files in the comfyui/models/vae directory, and rename it to flux_ae. Getting Started: Your First ComfyUI Apr 11, 2025 · Restart or Refresh ComfyUI. Apr 11, 2025 · Using Embedding Models in ComfyUI. If not, the defaults are probably fine. example, rename it to extra_model_paths. safetensors to make things more clear. On some model platforms, we can find embedding models that can output specific styles. That's one of the reasons I hate doing it manually off 1 folder or do it reverse and you'll end up erasing your own directory or hosting it in the wrong area. Nov 26, 2024 · What I did to correct the problem was add the directory of the LLM’s to the YAML. Use the filter to narrow down the model type. Place LoRAs in the folder ComfyUI/models/loras. By following this guide, you’ll learn how to expand ComfyUI’s capabilities and enhance your AI image generation workflow with these essential models. This guide will walk you through installing and running SDXL models in ComfyUI, a powerful open-source interface for AI image generation. I had a couple of custom bbox models and they didn’t show up. Visit the Hugging face model page and locate the specific model file you want to download; Right-click the 'Download' icon and select 'Copy link address' Apr 11, 2025 · Upscale Model Examples. Right-click on an empty space and select Add Node > image > upscaling > Upscale Image (using Model) to add the new node. yaml instead of . Jan 6, 2025 · To add model checkpoints, navigate to the ComfyUI ‘models/checkpoints’ folder and create subfolders for different model versions if necessary. type Since Loras are a patch on the model weights they can also be merged into the model: You can also subtract models weights and add them like in this example used to create an inpaint model from a non inpaint model with the formula: (inpaint_model - base_model) * 1. Using ControlNet Models. Introduction to LTX Video Model. Apr 11, 2025 · Since ComfyUI’s ecosystem is very rich, you may need to spend some time understanding the characteristics of different types of models and how to build various workflows. This guide provides detailed instructions on how to install and use Embeddings (Textual Inversion) models in ComfyUI, including downloading models, installation steps, and verification of successful installation, helping users quickly set up their AI image generation environment. 0, the base model will use the same weights as the model specified here (target model), and same for the clip with a corresponding strength_clip. yaml, then edit the relevant lines and restart Comfy. And above all, BE NICE. Step 3: Download models. In ComfyUI Desktop version, you can find extra_model_paths. Apr 11, 2025 · Learn about the ModelMergeAdd node in ComfyUI, which is designed for merging two models by adding key patches from one model to another. You can easily add or adjust models and configurations: Adding Models: Add a custom model to the /models/checkpoints/ directory by modifying the Dockerfile. I just copied and pasted the directory URL of A1111 folders into that, quite intuitive if done in an editor that allow to see the color of the text, like notepad++, thanks If you just want your likeness on models then there's reactor for face swapping, alternatively combining a faceid chained with a face plus ipadaptor gives very good results but is not a "true" face swap. This detailed step-by-step guide places spec Jan 15, 2024 · Click that text at the bottom and select the SDXL 1. 5-Model Name”, or do not rename, and create a new folder in the corresponding model directory, named after the major model version such as “SD1. Install models using ComfyUI Manager. 7. Download the ControlNet XL depth model. 5 is the latest AI image generation model, offering multiple powerful model variants. clip_name2: COMBO[STRING] Specifies the name of the second CLIP model to be loaded. bin. Here is an example: You can load this image in ComfyUI to get the workflow. Mar 14, 2023 · Also in the extra_model_paths. 5. Jan 1, 2025 · Here’s the download link for the DreamShaper 8 model. The Wan2. 2 Model File Structure. Welcome to the unofficial ComfyUI subreddit. ComfyUI. Feb 8, 2025 · Download the Flux. You can keep them in the same location and just tell ComfyUI where to find them. 5 model. Dec 26, 2024 · ricperry added the Feature A new feature to add to ComfyUI. Adding ControlNets into the mix allows you to condition a prompt so you can have pinpoint accuracy on the pose of Feb 1, 2024 · Ah, ComfyUI SDXL model merging for AI-generated art! That's exciting! Merging different Stable Diffusion models opens up a vast playground for creative exploration. Basically, at strength_model=1. 5”, and then copy your model files to “ComfyUI_windows_portable\ComfyUI\models\embeddings\SD1. Therefore, I recommend creating specific folders when saving models to identify the version, making selection easier: Download, browse and delete models in ComfyUI. They’re essentially pre-trained models that you can add to your setup, helping to create CivitAI open in new window - A vast collection of community-created models; HuggingFace open in new window - Home to numerous official and fine-tuned models; Download your chosen model checkpoint and place it in the models/checkpoints directory (create it if needed). yaml. ComfyUI is a node-based interface and inference engine for generative AI; Users can combine various AI models and operations through nodes to achieve highly customizable and controllable content generation Nov 23, 2024 · The easiest way to update ComfyUI is through the ComfyUI Manager. Right click and Navigate to: Add Node > sampling > KSampler A comprehensive ComfyUI installation guide, covering the software itself, custom nodes, plugins, and drawing models. Put it in the folder ComfyUI ComfyUI_windows_portable\\ComfyUI\models use "directory symbolic link" destination is the comfyui folder itself that will be replaced and source is the folder of where your models are. Within the ComfyUI manager, you’ll want to install the ComfyUI-GGUF Apr 11, 2025 · LTX Video Workflow Step-by-Step Guide. 6. This parameter enables the loading of a second distinct CLIP model for comparative or integrative analysis alongside the first model. Installing models in Google Drive. 5 Checkpoint model at step 1; Load the input image at step 2; Load the OpenPose ControlNet model at step 3; Load the Lineart ControlNet model at step 4; Use Queue or the shortcut Ctrl+Enter to run the workflow for image generation; Scenarios for Combining Welcome to the unofficial ComfyUI subreddit. Dec 30, 2024 · Update ComfyUI: Ensure you’re using the latest version of ComfyUI, as updates often include support for more model types. Just use ComfyUI Manger ! And ComfyAnonymous confessed to changing the name, "Note that I renamed diffusion_pytorch_model. Drag a model thumbnail onto an existing node to set the input field. 5 Large Turbo checkpoint model. The core node is the LoRA Loader node, which is responsible for loading LoRA models. Not sure where it is storing the Ultralytics models now. The YAML is located: C:\Users(USER_NAME)\AppData\Roaming\ComfyUI\extra_models_config. The primary example of a model data file in ComfyUI is an AI diffusion model. In this guide, we are aiming to collect a list of 10 cool ComfyUI workflows that you can simply download and try out for yourself. Most ComfyUI installations don’t include base models by default. bat to start ComfyUI. Users have the ability to assemble a workflow for image generation by linking various blocks, referred to as nodes. py --directml Embark on an intriguing exploration of ComfyUI and master the art of working with style models from ground zero. Feb 21, 2025 · 4. Here is an example of how to use upscale models like ESRGAN. safetensors or . Finally, add a “Save Image” node connected to the VAE Decode. yaml Model Library: M: Browse all locally indexed ComfyUI models, drag and drop to load models with corresponding loader nodes: 4: Workflows: W: Browse all your locally saved ComfyUI workflows, double-click to load a workflow in a new workflow tab window: 5: Theme Switch-Switch current ComfyUI interface theme, currently supports: Dark, Light: 6 Yeah, that did it, its the extra_model_paths. Apr 11, 2025 · 4. This tutorial provides a basic guide for using multiple LoRA models in ComfyUI, using the SD1. Oct 1, 2021 · ComfyUI is a versatile platform that allows users to enhance their imagery generation experience by integrating various image diffusion models. Connect it to a “KSampler” node. example (text) file, then saving it as . The first node you’ll need is the KSampler. Here is my way of merging BASE models and applying LORAs to them in non-conflicting way using the ComfyUI (grab the workflow itself in the attachment to this article): Nov 1, 2024 · These models are stored under the /models directory in the container. ip-adapter_instant_id_sdxl. Select the corresponding menu item and open it with a text editor. Alternatively, clone/download the entire huggingface repo to ComfyUI/models/diffusers and use the MiaoBi diffusers loader. My folders for Stable Diffusion have gotten extremely huge. ulqpbsno pwcdmr qpaj ettt agyxv gxzssv yckiw xfwl efwofn yoywl tagrzlkb eair desplm ixxpgz ocjj