ComfyUI Custom Extensions

WAI's Character select ported to ComfyUI with a few tweaks

ComfyUI custom node implementing the tile_colorfix color correction algorithm from A1111's ControlNet extension, supporting both SD/SDXL and Flux models. (Description by CC)

A soft, Anima-friendly Perturbed Attention Guidance node for ComfyUI.

A ComfyUI node that uses NVIDIA's DAM model to identify objects in masked regions

ComfyUI wrapper nodes for HunyuanVideo-Foley: Generate audio from video + text prompts

Combine multiple 3D models into a single Blender file and render in ComfyUI.

ComfyUI custom nodes for image workflows, tiling, masks, smart resize, logic routing, prompt and text tools, batch automation, garment/portrait segmentation, and VTON preprocessing.

kokoro-onnx (opensource TTS model) wrapper for ComfyUI.

Native ComfyUI implementation of the JoyAI-Image family supporting single and multi-image image editing and text-to-image generation using JoyAI-Image models with Qwen3-VL support.

Custom nodes to integrate the USO image generation model (based on FLUX.1 by ByteDance) into ComfyUI. This project is an implementation based on the official source code from the [a/ByteDance/USO](https://github.com/bytedance/USO) repository.

A custom ComfyUI node for visual comparison of two images using multiple perceptual and mathematical methods. The goal: make hidden differences visible as colorful heatmaps, so you can see where an upscaler, denoiser, or diffusion model changed your image — even when your eyes can’t tell at first glance.

Simple ComfyUI wrapper around DiffSynth-Studio's Z-Image i2L v2 (Image-to-LoRA).

Adds a custom node for saving images in webp, jpeg, avif, jxl (no metadata) and supports loading workflows from saved images

Float project applicable to ComfyUI.Generates speaking portrait video frames from an image and audio.

ComfyUI custom node pack for WAN v2v long video generation. Provides nodes for chunking, overlap stitching, control group orchestration, debug numbering, and de-overlap merging to create seamless long-form videos.

A custom ComfyUI node designed for stable, high-resolution video export — even on limited hardware.

Contains SlopPrompt and BooruPromptGenerator nodes to automate prompt generation for anime models. SlopPrompt allows natural language prompt generation or advanced formats like json and various conversions between them, while BooruPromptGenerator will let you generate random prompts anchored around tags you specify with various sampling options.

Using pyramid_noise instead of original noise in inference

A mobile-friendly drawer-style gadget platform for ComfyUI with Deck, Gallery, Model Viewer, XYZ Plot, and utility nodes.

Advanced multi-shot chaining for MiniMax-H3 audio+video with resumable cache, continuity controls, and unified model loading.

This is an experimental node for generating an *imatrix* file to reduce quantization errors in GGUF files used with ComfyUI-GGUF.

Automatically detect and download missing LoRAs for ComfyUI workflows

Full-canvas file explorer for browsing and managing images across ComfyUI's input, output, temp and arbitrary paths — thumbnails plus delete, rename and move.

ComfyUI nodes to use [a/1Prompt1Story](https://github.com/byliutao/1Prompt1Story)

Adding TCD sampling

Boost ComfyUI's web UI loading with object caching and gzip compression

ComfyUI nodes to use ReNoise

A collection of workflow helper nodes focused on organization + control (bypass/mute/collapse/group control), prompt/text utilities, file/path helpers, and image/video save helpers (optional VHS integration).

Add invisible watermark to images to protect your images

The ultimate visual browser for your local LoRAs and models! Enjoy browsing a stunning Visual Model Library with customizable thumbnails for your .safetensors files. Instead of struggling with a dead wall of text, you no longer have to endure the frustration of scrolling through endless lists of cryptic filenames in default loaders. Trying to memorize the style behind every file kills your creative flow. By eliminating guesswork, you get an intuitive feeling for your local collection. Instantly view beautiful preview images, read descriptions, and copy essential trigger words with a click. Stay perfectly organized with lightning-fast filtering, deep search, and custom personal notes for every model. Peace of mind for creators is guaranteed since our separated SFW and NSFW views ensure a safe interface for live streaming. This suite provides a Checkpoint, Diffusion, and LoRA Loader, all brought to vibrant life by a deep, automatic Civitai integration. Experience all your models!

一键把当前 ComfyUI 工作流推到 Modal Serverless GPU(H100)跑,自动同步本地模型/自定义节点,出图回填画板,零终端。 | One-click run the current ComfyUI workflow on Modal Serverless GPU (H100); auto-sync local models/custom-nodes, results back to the canvas, no terminal.

MagoStudio Nodes for ComfyUI

A handful of image filters for ComfyUI (darken, lighten, levels, saturate, hue)

Create colorful mosaic images in ComfyUI by computing label images and applying lookup tables.

This custom node for ComfyUI enables detailed segmentation of colored mask images into specific anatomical regions for male and female bodies.

Image obfuscation in ComfyUI

Essential utility nodes for ComfyUI including prompt queue, batch processing, camera angle control, and resolution selector. (Description by CC)

This platform extension provides ZhipuAI nodes, enabling you to configure a workflow for online video generation.

Sleek, modern resource monitors (CPU / RAM / GPU / VRAM / Temp / Disk) for the ComfyUI toolbar, with live sparklines.

Unattended multi-shot MiniMax H3 prompt chains for ComfyUI

With the MiaoShua creator's end, you can use this plugin to upload debugged workflows for consumer end users to use.

One hardware-aware loader for the whole MiniMax H3 Director model stack — model, CLIP, VAEs, LoRAs, and attention patches — auto-selected from your detected VRAM. Routes model loading through an adaptive VRAM reservation system that measures real usage per render shape and tightens itself over repeat runs, rather than a fixed guess.

ComfyUI node wrapping omnivoice-triton for Triton-fused OmniVoice TTS

A ComfyUI node that integrates LibLib's text-to-image and image-to-image generation capabilities, with customizable ControlNet and Lora support.

High-Performance Zero-OOM Multi-GPU Distribution & Memory Management Engine for ComfyUI — Layer Sharding, Device Splitting, and Dynamic VRAM Balancing for FLUX.2, LTX-Video, and Large DiT Models.

SegGPT model for comfyui,segmentation everything with mask prompt. Download (https://huggingface.co/BAAI/SegGPT/blob/main/seggpt_vit_large.pth) in this node path.

Custom utility nodes for ComfyUI by Nilor Corp. Probably not useful for most people, but contains stuff for working with lists, filenames, image batches, etc in a very specifc way.

125+ production-ready V3 custom nodes for ComfyUI covering file ops, S3 cloud storage, image/video/audio processing, model management (HuggingFace/Civitai/S3), OpenAI & Runpod integration, and workflow utilities.

Typed decisions (noul / choice / score) with calibrated probabilities, on ComfyUI's text-encoder engine. Currently imajev.

Drop in a generated image and recover its full ComfyUI workflow lineage — prompts, models, LoRAs, seed, sampler, and custom nodes. Local-first and privacy-preserving.

A plugin that enables one-click language switching between Chinese and English in ComfyUI without page refresh. (Description by CC)

A standalone custom node that creates a Prometheus-compatible metrics endpoint within the ComfyUI installation.

An interactive, classic channel mixer color adjustment node for ComfyUI, with realtime preview and a responsive editing interface.

This extension helps generate images through Illustrious.

ComfyUI EasyControl Nodes is a collection of nodes for ComfyUI that allows you to load and use EasyControl models.

Custom ComfyUI nodes including LLM integration, LoRA switchers, image tools, and more

Wrapper for the Sa2VA model

A text overlay node for ComfyUI that supports rich effects such as gradients, outlines, and shadows. It is suitable for adding highly customizable text content to images.

Use ngrok to allow external access to ComfyUI. NOTE: Need to manually modify a token inside the __init__.py file.

Warning, uses experimental package `comfy-env` to attempt a one click isolated install. Will download and use pixi package manager. Tiny Gaussian Splat viewer node for ComfyUI.

Warning, uses experimental package `comfy-env` to attempt a one click isolated install. Will download and use pixi package manager. ComfyUI utility pack for 360° equirectangular panoramas.

Tiny Gaussian Splat turntable viewer node for ComfyUI.

A custom node designed for ComfyUI, allowing users to format the current date and time based on a specified format.

Generates PBR textures

ComfyUI nodes that support SD3/SD3.5 in FlowEdit

Converts booru tags to a format suitable for Waifu Diffusion(or Danbooru based models).

Auto Prompt Schedule is a ComfyUI helper node for the Prompt Schedule Node or the Batch Prompt Schedule node from Fizzledorf. You need this Prompt Schedule nodes to make use of the Auto Prompt Schedule Node.

A ComfyUI extension that helps users relink missing models in workflows

A custom node for ComfyUI that calculates CLIP and LPIPS similarity scores between two images.

ComfyUI_Appstore, a tool that converts ComfyUI workflows into web apps on huaxiaobao.net with one click, and supports payments, like ComfyUI_Bxb (Bxb) does. Providing a way for the comfyui authors to get profit from.

A custom node for ComfyUI to save image sequences as video files using FFmpeg. Supports various codecs, audio muxing, and in-node previews.

Ultra-lightweight, high-fidelity text-to-speech for ComfyUI using Soprano TTS.

This is an implementation of [Qwen3-VL](https://github.com/QwenLM/Qwen3-VL) for [ComfyUI](https://github.com/comfyanonymous/ComfyUI), which supports for text-based and single-image queries.

Implementation of [a/Guidance in the Frequency Domain Enables High-Fidelity Sampling at Low CFG Scales](https://arxiv.org/abs/2506.19713) for ComfyUI.

Use the fewest nodes for the most flexible model calls. A versatile ComfyUI plugin for OpenAI-compatible APIs, featuring multi-purpose nodes for text and image, support for switching between multiple API providers, and auto-saving of custom models. Compatible with new models like nano-banana and seedream4.

Nodes:Edit Mask

Auto block-index remapping for applying old-Anima LoRAs and merging models with Anima-2.9B (28-to-40-block layer expansion)

Save and load images with prompts and generation metadata. Batch image loading with metadata extraction for ComfyUI

LongCat_Avatar,an upgraded open-source framework for audio-driven human video generation.

SteadyDancer: Harmonized and Coherent Human Image Animation with First-Frame Preservation,Minimize the number of libraries and VRAMs as much as possible

Compare one to nine images when use ComfyUI

Storyboard Suite — pre-production ноды для ComfyUI: таблица промптов и сетка кадров. Карта-ассет канала @itsmyshnikov

Adds a Navigator sidebar for keyboard-first workflow search and jump.

3D editor node — build a blockout scene (primitives, GLB, FBX with animation, gaussian splats), record a keyframed camera path, and feed the rendered mp4 + image + prompt into Seedance / Nano Banana / Flux / any downstream model.

This node is used to enhance image details. We can add a latent space image and introduce any amount of noise. Then, we can start denoising at any timestep. This allows us to add more details to the image while maintaining overall consistency as much as possible.

Nodes to use Florence2 VLM for image vision tasks: object detection, captioning, segmentation and ocr

A comprehensive ComfyUI node package that uses Microsoft's TRELLIS2 to transform video footage into 3D mesh reconstructions with advanced rendering effects.

Professional ComfyUI plugin for Sunra.ai's FLUX.1 Kontext and Seedance models with enhanced UI

NODES:Directory Loader, Filter Images Without Faces, Detect Faces and Draw Detection Box

A collection of ComfyUI custom nodes that integrate with MiniMax API services.

Native ComfyUI V3 nodes for DreamX-Creator joint audio-video generation and causal 2x refinement.

Standalone LLaDA-Image Base and Turbo AIO models with native ComfyUI model management, text/VQ generation and editing.

Four-step PDD sampler for Qwen-Image-2.1 Fun-Acc with stage-specific output heads.

This is an extension for ComfyUI. It retains multiple workflow tabs so that they are not lost when reloading or restarting.

Instead of LoraLoader or HypernetworkLoader, it receives a prompt and loads and applies LoRA or HN based on the specifications within the prompt. The main purpose of this custom node is to allow changes without reconnecting the LoraLoader node when the prompt is randomly altered, etc.

Displays a large draggable floating window with live denoising step previews as your image generates, similar to A1111 and Forge.

Add proper multi-GPU support to your MiniMax H3 workflows with these drop-in replacements for common nodes, that unlock parallel processing via SGLang.

custom node for ComfyUI that allows you to load prompts directly from CSV files. It is specifically designed for Bulk/Batch Image Generation, allowing you to automate the creation of hundreds of images using a pre-defined list of prompts.