# ComfyUI

LanPaint is an image inpainting plugin for stable diffusion models. Through multiple rounds of iterative inference, it can achieve high-quality image inpainting without additional training. The importance of this technology lies in its provision of a solution that allows users to obtain accurate inpainting results without complex training, greatly reducing the barrier to entry. LanPaint is suitable for any stable diffusion model, including user-defined models, and has wide applicability and flexibility. It is mainly aimed at creators and developers who need high-quality image inpainting, especially those who want to get quick inpainting results without additional training.

ComfyUI-WanVideoWrapper provides ComfyUI nodes for WanVideo, enabling users to leverage WanVideo's functionality within the ComfyUI environment for video generation and processing. Developed in Python, this tool supports efficient content creation and video generation, suitable for users needing to quickly produce video content.

ComfyUI-Hunyuan3DWrapper is a plugin based on ComfyUI that encapsulates the Hunyuan3D-2 model for efficient 3D image generation and texture processing. By streamlining the usage process of the Hunyuan3D-2 model, this tool allows users to quickly create high-quality 3D models and textures within the ComfyUI environment. It supports custom configurations and extensions, making it ideal for users engaged in efficient 3D content creation.

ComfyUI-MMAudio is a plugin based on ComfyUI that allows users to process audio using the MMAudio model. The main advantage of this plugin is its ability to deliver high-quality audio generation and processing capabilities, supporting various audio models and easily integrating into existing audio processing workflows. It is developed by kijai and is open source, available on GitHub. Currently, it is primarily aimed at tech enthusiasts and audio processing professionals and is available for free.

ComfyUI-MochiEdit is a video editing plugin based on Genmo Mochi technology, allowing users to edit videos through the ComfyUI interface. The primary advantage of this plugin is its ability to leverage advanced video processing techniques, providing users with an intuitive and user-friendly editing environment. Background information indicates it was co-developed by logtd and kijai and follows the GPL-3.0 open-source license. Due to its open-source nature, this plugin is available for free and is aimed at professional users or enthusiasts in need of video editing capabilities.

SDXL_EcomID_ComfyUI is a plugin that provides native support for SDXL-EcomID in ComfyUI. It enhances portrait representation, resulting in a more realistic and aesthetically pleasing appearance while ensuring semantic consistency and greater internal ID similarity. This plugin is fully integrated into ComfyUI and implements EcomID locally without relying on diffusers. Its significance lies in its ability to elevate the quality and consistency of image generation, particularly when working with portraits, ensuring consistent internal features despite physical changes such as age, hairstyle, and eyeglasses.

ComfyUI-Detail-Daemon is a node ported from muerrilla's sd-webui-Detail-Daemon for use within ComfyUI, allowing adjustment of sigma values that control detail. This tool is particularly suited for enhancing the details in the Flux model, while also potentially removing unwanted background blur. It includes four nodes: Detail Daemon Sampler, Detail Daemon Graph Sigmas, Multiply Sigmas, and Lying Sigma Sampler, offering multiple methods for enhancing image details.

EveryoneNobel is a framework that leverages ComfyUI to create personalized Nobel Prize images. It can be used not only for generating Nobel Prize images but also as a general framework to convert visuals produced by ComfyUI into final products, offering a structured approach for further applications and customization. This project demonstrates how to build a complete application and sell products within 30 hours, providing detailed installation and usage guides suitable for users looking to quickly generate personalized images.

ComfyUI-Fluxtapoz is a collection of nodes designed for editing images in ComfyUI using Flux. It enables users to edit images and apply style transformations through a series of node operations, making it particularly suitable for professionals engaged in image processing and creative work. This project is currently open-source and follows the GPL-3.0 license, meaning users can freely use, modify, and distribute the software, provided they adhere to the terms of the open-source license.

ComfyUI LLM Party aims to create a complete set of LLM workflow nodes based on the ComfyUI frontend, enabling users to quickly and easily construct their own LLM workflows and seamlessly integrate them into existing image workflows.

ComfyUI-GGUF is a project that provides GGUF quantization support for native ComfyUI models. It allows model files to be stored in GGUF format, which is promoted by llama.cpp. Although standard UNET models (conv2d) are not suitable for quantization, transformer/DiT models like flux appear to be minimally affected by quantization. This enables them to run on low-end GPUs with lower bits per weight variable rate.

The SaltAI Language Toolkit is a project that integrates the retrieval-augmented generation (RAG) tool Llama-Index, Microsoft's AutoGen, and LlaVA-Next. It enhances the functionality and user experience of the platform through the adaptable node interface of ComfyUI. The proxy function was added on May 9th, 2024.

BizyAir, developed by siliconflow, is a plugin designed to help users overcome environmental and hardware limitations, making it easier to generate high-quality content with ComfyUI. It supports running in any environment, eliminating concerns about environmental or hardware requirements.

ComfyUI_omost is an Omost model implemented within the ComfyUI framework. It enables users to interact with large language models (LLMs) to obtain JSON-like structured layout prompts. This model is currently under development, and its node structure may be subject to change. It translates JSON conditions into ComfyUI's region format for image generation and editing through two parts: LLM Chat and Region Condition.

cog-consistent-character is an AI-powered image generation model that lets users create images of a given character in various poses. Leveraging Stable Diffusion technology and offering a user-friendly interface through ComfyUI, it enables even those without programming experience to easily generate high-quality images.

AnyNode is a plugin designed for ComfyUI. It leverages the capabilities of LLMs (large language models) to generate the desired output based on user input. It supports the use of OpenAI API or local LLMs API, allowing users to achieve complex programming tasks through simple configuration and instructions without writing code. The key advantages of this plugin include ease of use, flexibility, and powerful functionality, which can significantly improve development efficiency, especially for developers who need rapid prototyping and automation tasks.

ComfyUI ProPainter Nodes is a video repair plugin based on the ProPainter framework. It uses flow propagation and spacetime transformers to achieve advanced video frame editing, ideal for seamless repair tasks. The plugin features a user-friendly interface and powerful functionalities designed to simplify the video repair process.

ComfyFlow is a workflow application creation platform built on top of ComfyUI, enabling you to quickly build and share workflow applications with others. Leveraging the power of Stable Diffusion and ComfyUI technology, it offers simplicity, full hosting, and free usage.

MagicTime merges the MagicTime time Lora into the Animatediff v3 dynamic model and converts its spatial Lora to .safetensors format. Use it together with ComfyUI and AnimatediffEvolved. Powerful and capable of providing high-quality spatiotemporal fusion effects, specifically designed for dynamic model applications. Pricing is determined based on specific usage and focuses on providing advanced spatiotemporal fusion technology.

ComfyUI-Gemini is a plugin that incorporates the Google Gemini model into ComfyUI. Users can leverage Gemini to generate prompts, chat with it, and utilize multimodal input including images. This plugin is free to use, offering both implicit and explicit API key usage methods, suitable for both individuals and teams.

This is a reference implementation of the ComfyUI IPAdapter model. IPAdapter is a powerful model for image-to-image conditional generation based on one or more reference images. Through text prompts, control networks, and masks, you can generate variations of enhanced images. It can be seen as a Lora for a single image. This implementation is memory-efficient, runs quickly, and will not be disrupted by Comfy updates. As an open-source project, developers are welcome to contribute to support project maintenance and new feature development.

This is a custom sampler plugin node for ComfyUI, implementing the trajectory consistency distillation (TCD) sampling method proposed by Zheng et al. The plugin adds TCDScheduler and SamplerTCD nodes to the ComfyUI's Custom Sampler category. Simply clone it to the custom_nodes folder and restart ComfyUI to use it. TCDScheduler has a special parameter `eta` that controls the randomness at each step. When `eta`=0, it represents deterministic sampling, and when `eta`=1, it represents completely random sampling. The default value is 0.3, but a higher `eta` value is recommended when increasing the inference steps. This plugin, based on the trajectory consistency distillation sampling method, can provide smoother and more consistent output results for AI models.

ComfyUI-3D-Pack is a powerful collection of 3D processing plugins that extend ComfyUI's capabilities to handle 3D models (meshes, textures, etc.), incorporating cutting-edge 3D reconstruction and rendering algorithms such as 3D Gaussian sampling and differentiable NeRF rendering. It enables fast reconstruction of 3D Gaussian models from single-view images, which can then be converted into triangular mesh models. Additionally, it provides an interactive 3D visualization interface.

The ComfyUI-3D-Pack is a powerful node plugin for 3D processing, enhancing ComfyUI with the capability to handle 3D inputs (meshes, UV textures, etc.) using cutting-edge algorithms such as 3D Gaussian sampling and neural radiance fields. This project allows users to rapidly generate 3D Gaussian models from a single image, converting Gaussian models to meshes for 3D reconstruction. It also supports multi-view images as inputs, enabling the mapping of multi-view rendered textures onto the given mesh UVs. Currently in development and not officially released to the ComfyUI plugin library, it already includes features such as large multi-view Gaussian models, three-plane Gaussian transformers, 3D Gaussian sampling, depth mesh triangulation, and 3D file loading and saving (obj, ply, glb, etc.). Its aim is to become a robust tool for ComfyUI's 3D content processing.

Comfy Textures is a Unreal Engine plugin that integrates the editor with ComfyUI, allowing for the quick creation and adjustment of scene textures using generative diffusion models. Supports single-viewpoint and multi-viewpoint texture projection and is suitable for both perspective and orthographic cameras. It also offers texture editing and image-to-image workflows. It can seamlessly work with Unreal Engine versions 5.x and 4.x.

This is an unofficial implementation of InstantID within the ComfyUI environment, allowing users to experience the character image generation effects brought by InstantID without deploying a server. It supports automatic model download from huggingface hub and can also load local models. Compatible with various text prompt inputs and styler, it can easily realize different styles of character image generation.

ComfyUI-Moore-AnimateAnyone is a ComfyUI custom node based on the Moore-AnimateAnyone model. It can generate corresponding human animations through simple text descriptions. The node is easy to install and use, supports the generation of various human postures and movements, and can be used to enhance the quality of design works. Its output animation is delicate and natural, providing creators with a powerful tool.

Plush-for-ComfyUI is a plugin that offers custom nodes for ComfyUI, allowing the generation of TextNodes from either text or images, thereby expanding ComfyUI's support for varied prompt input methods. The plugin features an interactive node-connection interface, supports models gpT-3 and gpT-4, and allows customization of the degree of creativity in prompt generation, supports multiple artistic styles, and can provide reference information on selected styles.