ComfyUI Custom Extensions

A mask merge/edit node that merges masks from a mask batch into one mask as output for future use, using add mode.

ComfyUI custom node for GPT Image2 text-to-image and image-to-image generation through OpenAI-compatible endpoints.

Simple node for sharing latent image size between nodes. Preset dimensions for SD and XL.

A simple markdown node for ComfyUI

A collection of useful nodes for ComfyUI

A collection of useful nodes for ComfyUI, including Float Preview for live image visualization.

Independent Hrio Design custom nodes for ComfyUI. Graphic design and interior design prompt editor, single image node, single video node, and three-variant design generation.

ComfyUI custom nodes for Krea2-style multi-reference image editing.

A premium ComfyUI custom node for image denoising with 6 algorithms, dark neon-themed UI, and built-in sharpening.

WebApp API library, A collection of custom nodes to define a web app interface for your workflows. WebApp frameworks can use the API to generate apps and connect with the workflow.

A ComfyUI node to enhance the security of the overall ComfyUI experience, for both private and public-facing servers, adding more features, while still allowing ComfyUI to be usable.

This repository encapsulates the BAGEL model as ComfyUI nodes for use, including image editing and image inversion features, but it does not include text-to-image functionality.

An extension set of validation checks, automatic versioning numbering, automatic directory creation, and naming conventions are implemented to ensure that the file system is kept in sync with ShotGrid.

A simple, no-node-graph way to generate images with the Anima model in ComfyUI, with built-in LoRA and style management.

This repository provides two powerful custom nodes for ComfyUI to create stunning before-and-after transition videos. These nodes are designed for visual comparisons, transformations, and creative effects, supporting both standard and depth map-based transitions.

Fork of muse-ltx-timeline (V1) adding Seed Hunt: a 4-seed Stage-1-resolution scouting pass reusing the real timeline's guide/audio data — pick a candidate visually and its actual latent carries forward into Stage 2 (upscale + refine), instead of regenerating from scratch. V1 is included untouched alongside it.

Standalone img2img-style second-pass 'hi-res fix' node for MiniMax H3 — pick one of up to four candidate clips (e.g. cheap low-res scouting passes) and re-sample it at a higher resolution via partial-denoise, with optional reference-image identity locking. Companion to Muse Minimax Director, no dependency on it.

Swap the speaking voice in a mixed audio clip (dialogue + ambient sound) while keeping the background intact — Demucs separation, Whisper transcription, local-Gemma emotion tagging, ECAPA-TDNN speaker auto-assignment, Fish Audio S2 cloning, with a Muse-Director-style timeline editor for reviewing/correcting segments.

Custom nodes for ComfyUI that extract text prompts from PNG image metadata

markdown file selector for credentials and system prompts on demand

ComfyUI nodes for loading and applying VST and Audio Unit plugins to audio within ComfyUI workflows.

Chinese Traditional and Simplified converter

A custom node that returns the generation time of the KSampler. Intended for benchmarking or debugging.

ComfyUI custom nodes for generating original character prompts and reusable settings.

A collection of custom nodes for ComfyUI that send video data through websockets instead of saving to disk.

workflow.json -> workflow_api.json

Passthrough nodes that free models ComfyUI's memory manager will not free, so large models fit on small cards.

NeoGriever's helper nodes. Better CLIP Text Encoder, Resolution Provider, Multimask Write/Read, TextBoxes Simple/Join/x2/x3, Sliders INT/FLOAT/STEPPER, String Tool/Squisher/Cutter, Create Solid Color, Fill with Color, Checkerboard Generator, Image Progress Bar

This custom ComfyUI node transforms a core idea into a richly detailed positive prompt using a local [a/Ollama](https://ollama.com) LLM.

#Camera Movement Prompt Node for ComfyUI This custom node script for ComfyUI generates descriptive camera movement prompts based on user-selected movement options for Wan2.2

Small custom node for ComfyUI that helps pick a base resolution for LTX workflows and shows the final resolution after two 2x upscale passes.

Save and load CONDITIONING tensors to or from files, enabling reuse of encoded prompts and caching expensive CLIP encodings across workflows.

Nodes: yugioh_Presets. by Nicolai256 inspired by throttlekitty SDXLAspectRatio

Frees VRAM memory when you are idle. Like the way ollama works.

Nodes: Aspect Ratio, Prompt Multiple Styles Selector, Text Concatenate

Nodes: DepthAnything v2, LineArt, PyraCanny, Sketch

All Credits go to the original Repo: [a/Hzzone/pytorch-openpose](https://github.com/Hzzone/pytorch-openpose).

A comprehensive performance monitoring and analysis tool for ComfyUI workflows that tracks execution time, memory usage, and system resource utilization.

Standalone Anima loader, sampler, prompt saver, and dynamic text utility with optional Easy-Use integration.

Full-reference image quality metrics for ComfyUI - PSNR, MAE, and SSIM between two images.

Utility collection of ComfyUI V3 nodes for stabilizing temporal segmentation masks while staying deterministic and torch-only.

ComfyUI custom node converting images to text prompts via external API, with automatic image preprocessing, base64 encoding, intelligent retry logic, and dual output modes. (Description by CC)

A floating dialog-based image preview and selection system for ComfyUI.

www.theoath.studio/projects/comfy-prompt-iterator?utm_source=registry.comfy A ComfyUI node that automatically steps through every line in a text file in a directory each generation run.

https://www.theoath.studio/projects/comfy-vllm-node Generates Stable Diffusion prompts using a locally running vLLM server. Supports {wild|card} expansion, a prefix field for quality tags, auto model detection via /v1/models, live preview on the node face, and configurable retry logic. No API keys or internet connection required. Wire the single output directly into CLIPTextEncode.

TikTok Caption — ComfyUI custom node that renders CSS-styled captions/subtitles from SRT or plain text using the tscaps engine, with a 1:1 in-node live preview.

Simple Notes

Automatically archives generated files to Amazon S3 or S3-compatible storage services

Dynamic block swap for Krea 2. Offloads transformer blocks to RAM to run on low-VRAM GPUs without quality loss.

A custom node for ComfyUI that integrates with OneThing AI's image generation API.

A set of custom nodes for ComfyUI that provides enhanced string manipulation and prompt variant generation functionality for AI workflows.

ComfyUI custom nodes adding asset resolution and publishing via OpenAssetIO

This project offers a collection of custom nodes for ComfyUI, dedicated to creating hexadecimal color palettes

Nodes to create D&D5E NPC character sheets

A video-first custom node package for converting flat video into 3D stereoscopic output inside ComfyUI using Depth Anything 3 AI model for depth estimation on monocular input video.

ComfyUI custom node pack for fal.ai image generation supporting FLUX models with queue polling, safety controls, and native image I/O conversion.

Browse and compare newly generated images, videos, and audio in ComfyUI, with batch views, metadata inspection, and favorites.

Nodes: Make Square Node, Interrogate Node, TextEncodeAIO

Nodes:Preview SVG

A lightweight, optimized custom guider node for ComfyUI that enables single-model mode for Ideogram 4, reducing VRAM usage and computational overhead.

ComfyUI node that converts <lora:name:weight> tags into Civitai-friendly Name:HASH:Weight metadata strings.

Browser-played workflow success and failure audio notifications for ComfyUI

A ComfyUI-native live resource monitor and snapshot node

Free trial of Tongyi Wanxiang wan2.1 model, this is a batch implementation of wan2.1 API, providing batch processing for your short video production.

Offline text-driven image editing for urban and architectural workflows using open-source Qwen-Image-Edit model in GGUF format with custom ComfyUI nodes for batch and stochastic image processing.

Gemini-powered image generation for ComfyUI supporting text-to-image, image-to-image, and batch processing with flexible API configuration. (Description by CC)

Offline translation node for ComfyUI, powered by Facebook's M2M-100 multilingual model. Translate text between 100+ languages — completely offline and private.

This extension adds cloud storage support to ComfyUI, allowing you to save and load images, masks, and videos directly from cloud storage services. Currently supports Aliyun OSS and AWS S3.

A custom ComfyUI node that compares face identity between two images using InsightFace embeddings. Useful for verifying that a generated image preserves the identity of a reference person.

Nodes:RandomPrompt, RandomPromptMixed, ImageScaleTo, EmptyLatentImageScaleBy, LoraLoaderExtended, Save Image w/Metadata, CheckpointLoaderSimpleExtended

A ComfyUI nodes package providing access to 300+ AI models through a unified Polza.ai API

Custom ComfyUI nodes for CorridorKey, a neural network green screen keyer that solves color unmixing and produces clean alpha channels from footage.

FlashVSR v1.1 video super-resolution built around ComfyUI's stock Wan implementation (compatible with Dynamic VRAM and int8 convrot). Mitigates OOMs on low-end hardware.

This is a custom node of ComfyUI that downloads and loads models from the input URL. The model is temporarily downloaded into memory and not saved to storage. This could be useful when trying out models or when using various models on machines with limited storage. Since the model is downloaded into memory, expect higher memory usage than usual.

Image data check, filtering and augmentation tools for ComfyUI 🔬 NOTE: Renamed from 'comfyUI-image-search'

Access llama.cpp server, plus ease-of-life functionality

ComfyUI native nodes for IC-Light with a Video node

ComfyUI Workflow Protector is a security plugin for ComfyUI that provides multi-layered protection against unauthorized export, copying, or usage of workflows. It supports password-based access control and workflow file encryption to safeguard your creative work.

ComfyUI custom node for OpenAI DALL-E 3 image generation and editing with support for multiple reference images and localized edits. (Description by CC)

ComfyUI custom_node that publish output image to rabbit_mq

Camera utilities for ComfyUI. Convert between camera formats (DA3 extrinsics/intrinsics to Load3DCamera for Preview3D).

Warning, uses experimental package `comfy-env` to attempt a one click isolated install. Will download and use pixi package manager. ComfyUI wrapper for HY-Pano 2.0 (Tencent): image -> 360° equirectangular panorama via Qwen-Image-Edit + HY-Pano LoRA.

Warning, uses experimental package `comfy-env` to attempt a one click isolated install. Will download and use pixi package manager. ComfyUI wrapper for HY-World 2.0 / WorldMirror 2.0: 3D reconstruction (depth, normals, point clouds, 3DGS) from multi-view images.

GPU particle studio that lives inside ComfyUI

A comprehensive collection of 18+ utility nodes for ComfyUI including image list operations (split/merge), advanced tiling with seamless blending, brightness/contrast/saturation/RGB adjustments, diagonal watermarking, overlapping image processing, image tile processing and various workflow optimization tools. Perfect for high-resolution image processing, batch workflows, and color grading.

ComfyUI custom node for portrait segmentation and face parsing using ONNX exports of the pretrained EasyPortrait checkpoints.

Aligns any input frame number to the nearest 4N+1 format, ensuring compatibility with video generation models that require specific frame count constraints.

NODES:AIRedoon Image Caption. Analyze image tagger

This Extension adds a few custom QOL nodes that ComfyUI lacks by default.

ComfyUI Batch Toolkit: Custom nodes that simplify batch operations and improve efficiency.

ComfyUI GPT Image 2 / 2.5 文生图与图生图节点，支持 GPT Image 2.5 Flare/Sunburst、透明背景、PNG/JPEG/WebP、压缩、自定义尺寸与最多16张参考图。

A ComfyUI custom node that renders a visual bounding-box / caption editor on the node itself and outputs the assembled Ideogram-4 caption (v15 format) as a JSON string.

Batch convert Anima 28-layer LoRAs to 40-layer key structure and save them into ComfyUI's loras folder.

This is a modified implementation of impact-pack's iterative upscaler. It leans in on the idea that giving too much attention to computation at high resolutions isn't a good idea.

A ComfyUI custom node that converts Civitai share/model page URLs into direct download URLs, making it easier to integrate Civitai models into your workflows.

A ComfyUI custom node that converts Civitai share/model page URLs into direct download URLs, making it easier to integrate Civitai models into your workflows.

Call the local `claude` CLI from ComfyUI for vision+text or text-only completions. Uses your existing `claude login` OAuth (no API key required). Supports UUID session reuse for prompt-cache savings, with session token-usage tracking. Additional utility nodes: `Split Image Batch to List of Image Batches` (takes a large incoming batch and splits it into a list of smaller sub-batches with start/end index outputs, so you can send N images at a time — e.g. pairs — to the Claude CLI node along with their original batch indices for tracking), `Indexed 2x2 Grid from Batch`, `Select Images (Bounds Safe)`, and `Image Batch Current Index`. Optional headless API-key mode.

A custom node extension for ComfyUI that enables conversion between different prompt formats: NovelAI V4, ComfyUI, and old NovelAI.

Added mask area.

A ComfyUI custom node that enhances image loading with subfolder organization and dynamic filtering