ComfyUI Custom Extensions

ComfyUI nodes to use [a/APG scaling](https://huggingface.co/papers/2410.02416) for CFG, allowing for better image quality with higher CFG.

Remove background of plural images.

Prompt editing, ANIMA prompt correction, NAIA prompt integration, LoRA preset management, wildcard expansion, AiO generation, and ANIMA/Spectrum workflow helpers for ComfyUI.

A ComfyUI extension for generating captions for your images. Runs on your own system, no external services used, no filter. Uses various VLMs with APIs to generate captions for images. You can give instructions or ask questions in natural language.

ComfyUI自定义节点，集成ByteDance Sa2VA模型，实现智能图像分割和视觉理解。

ComfyUI custom nodes for TencentARC Pixal3D image-to-3D generation and GLB export.

ComfyUI Web Extension for saving views and navigating graphs.

This is a node that outputs tracking results of a grid or specified points using CoTracker. It can be directly connected to the WanVideo ATI Tracks Node.

Nodes:sizing_node. Size calculation node related to image size in prompts supported by SDXL.

An image generation based music visualizer integrated into comfyanonymous/ComfyUI as custom nodes.

Turn images into seamless textures and build tiled previews. Unique radial mask method available.

flux lora merge. adaptive Merge (uses tensor norms and weight), manual Merge (uses fixed weights you specify), additive Merge (uses 100% of the first and adds a percentage of the second)

Make it possible to edit the prompt using the Monaco Editor, an editor implementation used in VSCode. NOTE: This extension supports both ComfyUI and A1111 simultaneously.

Ultimate high-performance hires fix solution combining latent/model upscaling, sampling, and advanced histogram-based color correction into one unified node.

Strict Sage 3 (Blackwell) attention for ComfyUI official flows.

ComfyUI-Merlin is a custom node extension for ComfyUI, introducing the Magic Photo Prompter. This powerful tool enhances your prompt engineering process by allowing users to easily construct detailed, high-quality prompts for photo-realistic image generation.

ComfyUI-HunyuanVideo-Avatar is now available in ComfyUI, HunyuanVideo-Avatar is a multimodal diffusion transformer (MM-DiT)-based model capable of simultaneously generating dynamic, emotion-controllable, and multi-character dialogue videos.

ComfyUI-ThinkSound is now available in ComfyUI, ThinkSound is a unified Any2Audio generation framework with flow matching guided by Chain-of-Thought (CoT) reasoning.

A ComfyUI custom node that uses an LLM agent to intelligently segment specific characters using SAM3

a comfyui custom node for [a/MARS5-TTS](https://github.com/Camb-ai/MARS5-TTS)

a custom node for [a/MiniMates](https://github.com/kleinlee/MiniMates)

WYSIWYG visual timeline director for MiniMax H3. Generates continuous ultra-long videos block-by-block with automated keyframe flow chaining and low, constant VRAM usage.

PromptModels Studio-nodos ComfyUI para NVIDIA NIM, Google Gemini y Grok. Zero SDKs externos, solo REST nativo.

ComfyUI-CameraCtrl-Wrapper

This is an implementation of [a/i2vgen-xl](https://github.com/ali-vilab/i2vgen-xl)

This is an ComfyUI implementation of RAFT to generate motion brush

A universal neural network latent upscaler that supports SD1.5, SDXL, Flux, and Wan2.2 models. Uses trained neural networks instead of simple interpolation for higher quality latent upscaling. Built upon the excellent foundation of [a/Ttl's ComfyUi_NNLatentUpscale](https://github.com/Ttl/ComfyUi_NNLatentUpscale) - this project extends the original work with universal model support and improved architectures.

Allows native usage of ModelScope based Text To Video Models in ComfyUI

NODES:Eagle Image Node for PNGInfo, SDXL Resolution Presets (ws), SDXL Prompt Styler, SDXL Prompt Styler Advanced

Adds video continuation / extension capabilities to Minimax-H3 in ComfyUI

The KimaraAIWatermarker custom node allows you to apply watermark text and logo overlays to images. Optionally, the watermark can be moved by the move_watermark_step amount of pixels after each generated image. To apply a moving watermark to a list of images, use the KimaraAIBatchImages node to concatenate the list into a single tensor, then use that as an input for the watermark node, as shown in the example image below.

Frame Selector & Sequence Selection Node for ComfyUI.

This repository is based on [https://github.com/BoyuanJiang/FitDiT](https://github.com/BoyuanJiang/FitDiT) and simply wraps it as nodes for ComfyUI.

Complete collection of image effects for ComfyUI - 32 nodes across 6 categories

A full prompt-writing studio in a single node: AI prompt enhance, variations, aesthetic inverter, and natural-language instruct editing driven by your local Ollama models. Includes style targeting with sliders, image to prompt and video to prompt captioning, a built-in browser for Civitai and Genur.art prompts and workflows, favorites and snippets, and wireless prompt injection into any text widget with no wires.

Warning, uses experimental package `comfy-env` to attempt a one click isolated install. Will download and use pixi package manager. CAD file processing nodes for ComfyUI with interactive 3D OCC CAD viewer. Load STEP, IGES, BREP files and generate meshes using OCC, GMSH.

Film Grain simulation, Log Color Conversions, Color Scopes (RGB Parade, Vectorscope, Gamut Warnings), False Color, and more.

Randomizes selected LoRAs and strengths. Includes trigger word output and support for exclusive/random selection.

A set of custom nodes for creating image grids, sequences, and batches in ComfyUI.

Load a folder of images one at a time, in order — a fresh image on every Queue. Natural sort, filetype picker, reverse, subfolders, hold/pause, scrub bar, folder browser, and current/next previews.

Speed by Simplicity: A Single-Stream Architecture for Fast Audio-Video Generative Foundation Model

Step_Audio_EditX：the first open-source LLM-based audio model excelling at expressive and iterative audio editing—encompassing emotion, speaking style, and paralinguistics—alongside robust zero-shot text-to-speech (TTS) capabilities，try it in comfyUI

UltraFlux:Data-Model Co-Design for High-quality Native 4K Text-to-Image Generation across Diverse Aspect Ratios,try it in comfyUI

The Qwen3 ComfyUI Integration is a powerful tool designed specifically for ComfyUI workflows, aiming to seamlessly integrate Qwen series large language models (LLMs).

Diffusion-Pipe In ComfyUI Custom Node is a powerful extension plugin that provides complete Diffusion model training and fine-tuning capabilities for ComfyUI. This project allows users to configure and launch training for various advanced AI models within ComfyUI's graphical interface, supporting both LoRA and full fine-tuning, covering the most popular image generation and video generation models available today.You can train Qwen lora with 16g Vram

Visualize and manipulate the latent space in ComfyUI

Node for saving PNGs, WEBPs, and JPEGs with rich automatically captured metadata from any node: prompts, model/VAE/LoRA names & hashes, sampler & scheduler (Civitai-compatible), optional workflow embedding, dynamic rule scanning + forced include system, deterministic ordering, filename tokens, and staged JPEG metadata fallback.

A ComfyUI plugin based on HunyuanVideo-1.5, offering both simplified and complete node sets for quick usage or deep workflow customization.

This node uses the RealESRGAN model from [a/xinntao](https://github.com/xinntao/Real-ESRGAN).

Partial redraw sampler and variant seed sampler

Suitable for Windows - MaskGCT ComfyUI Node Wrapping

a custom node for separation vocals from music based on [a/ZFTurbo/Music-Source-Separation-Training](https://github.com/ZFTurbo/Music-Source-Separation-Training)

Nodes:ImageLoader, ImageSaver, ImagesSaver, ImageResize, ImageSize, GFPGANNode, MaskAddNode, Video Load, ...

Custom nodes to run fal-ai/AuraSR model.

A ComfyUI custom node that brings Zonos Text-to-Speech capabilities to your workflows, featuring high-quality speech synthesis and voice cloning.

This contains all-in-one 'principled' nodes for T2I, I2I, refining, and scaling. Additionally it has many tools for directly manipulating the color of latents, high res fix math, and scripted image post-processing.

XiaoMi GemmaX: Support 28 languages, Multilingual Translator based on Gemma. A node for ComfyUI.

Experimenting with generating noise directly in the latent space.

TeaCache acceleration for the native ComfyUI implementation of circlestone-labs/Anima.

ComfyUI nodes for [MiVolo V2 Age & Gender Prediction](https://huggingface.co/iitolstykh/mivolo_v2)

Nodes:Essential Params, Image Params, Video Params, Add Margin With Color, Resize to megapixels

ComfyUI node to vectorize 2 color images like logo or text calling the pure Python 'potracer' library for potrace. Requires LATEST VERSION of https://registry.comfy.org/nodes/ComfyUI-ToSVG to_SVG node to save as SVG.

Ranbooru is an extension for the comfyUI. The purpose of this extension is to add a node that gets a random set of tags from boorus pictures. This is mostly being used to help me test my checkpoints on a large variety of

Gridswapper takes a batch of latents and spreads them over the necessary amount of grids. It then automatically shuffles the images in the grids for each step. So, a batch of 12 latents for a 2x2 grid will generate 3 grid images in each step. It will then shuffle around the images for the next step. This makes it possible for all images to influence the others during the denoising process. This approach works well for generating 2-4 grids.

All-in-one ComfyUI image generation node: switchable architecture (AuraFlow, Flux.1/2, Krea 2, Qwen-Image-Edit, SD1/SD1.5/SDXL, SD3/3.5), tri-source model loading, dual/triple CLIP, LoRA stack, prompt enhancer, refiner pass, optional upscale, A/B compare slider - in a single node.

Save Image with more file formats for ComfyUI

This node is mainly based on the Yolov8 model for object detection, and it outputs related images, masks, and JSON information.[w/Repository url is changed. Please remove previous one and reinstall.]

Experimental ComfyUI nodes for training in-memory Anima reference context tokens.

Advanced Text Overlay Node for ComfyUI featuring rebuilt rendering logic, precise alignment, custom fonts, and extended styling controls.

An unofficial ComfyUI custom node integration for High-quality Text-to-Speech and Voice Conversion nodes for ComfyUI using ResembleAI's ChatterboxTTS with unlimited text length!!!.

A comprehensive ComfyUI wrapper for HiggsAudio v2, enabling high-quality text-to-speech generation with advanced voice cloning capabilities. Supports multiple voice presets and custom reference audio for voice cloning. Requires transformers==4.45.2 for compatibility.

A ComfyUI production pack with Studio workflow cores, Outfit Forge, a portable Templates/Prompts/Outfits/Scenes Creative Library, a visual LoRA Library, and safe LoRA organization.

Scheduler for ComfyUI and an attempt at optimized scheduler for the Chroma architecture.

Llama3_8B for comfyUI， using pipeline workflow.

Nodes: ComfyUI_Mexx_Poster

Nodes: Image Resize (seam carving). Seam carving (image resize) for ComfyUI. Based on [a/https://github.com/li-plus/seam-carving](https://github.com/li-plus/seam-carving). With seam carving algorithm, the image could be intelligently resized while keeping the important contents undistorted. The carving process could be further guided, so that an object could be removed from the image without apparent artifacts.

Composable Breeze TTS 2 voice design, cloning, direction, and desktop voice-bundle nodes for ComfyUI

A logic engine for ComfyUI prompts that transforms static prompts into dynamic, context-aware workflows with persistent variables, conditional logic, native LoRA loading, and external data fetching. (Description by CC)

We developed a custom_node for Liveportrait_v3 that enables flexible use on Comfyui to drive image-based emoji generation from photos.

ComfyUI custom node port of Sprite Fusion Pixel Snapper that snaps pixel art to a clean grid and quantized palette.

Nodes:GPT4V-Image-Captioner

A custom node for a LMStudio integration into ComfyUI.

a plugin of ComfyUI for CosyVoice2

Sigma-aware model LoRA strength scheduling for ComfyUI

Custom workflow-oriented nodes for ComfyUI

A collection of custom nodes for ComfyUI focused on animation, image processing, and workflow optimization.

Block-weighted LoRA stacking with CivitAI metadata, hash resolution, and browser integration; Donut Detailer for per-block model tuning; TeaCache acceleration; tiled upscaling; CFG sampling with 18 curve types; and reference-based spectral sharpening.

VS Code-inspired MiniMax H3 prompt editor with strict prompt sections, completions, reference media previews, and plain STRING output.

LTX Director Motion Brush timeline tools for LTX 2.3 workflows.

Wrap Rope into ComfyUI, do a little change to use in ComfyUI. All credit goes to Hillobar and his ROPE [ㅁ/https://github.com/Hillobar/Rope](https://github.com/Hillobar/Rope)

A multimodal ComfyUI AI node with Qwen3.5, Qwen3-VL, Qwen2.5-VL, Qwen3, and Gemma 4 integrations. Features live thinking in the terminal to see what the LLM is doing in real time.

Dual implementation of reference-based video colorization featuring ColorMNet (2024) with DINOv2 and Deep Exemplar (2019). Includes 4 nodes (2 video, 2 image), multiple feature encoders (VGG19, DINOv2, CLIP), advanced post-processing (color-matcher, WLS, guided, bilateral), and auto-installer for dependencies.

Reference images for Krea 2 via native Qwen3-VL vision tokens: character and style transfer with the companion reference LoRA (layered loader included).

A suite of custom ComfyUI nodes for building real-time video and audio workflows using ComfyStream.

a ComfyUI plugin to transform APP Mode workflow to MCP

Optional wildcards in ComfyUI

Custom-ink halftone for ComfyUI. Pick any set of inks plus a paper colour and the node solves per pixel for the best mix of those inks, screens each plate at its own angle and stacks them in print order. 17 patterns, 37 presets, real Ben-Day plate rendering and tint calls.

This repository provides the official ComfyUI workflow for [a/Insert Anything](https://github.com/song-wensong/insert-anything).

A comprehensive node package that seamlessly integrates all BiRefNet series models into ComfyUI