ComfyUI Custom Extensions
Browse 9,446 ComfyUI extensions and 73,215 custom nodes.
This nodepack lets you use several Replicate models for image generation and editing: Qwen Image Edit Plus (Multi-Image) for editing multiple images using reference poses, Qwen Image Edit (Single Image) for single-image edits with text instructions, SeedEdit 3.0 (ByteDance) for advanced prompt-controlled editing, and Seedream 4.0 (ByteDance) for high-quality multi-resolution image generation.
This is an extension that automatically saves video files generated with ComfyUI's 'video combine' extension to the Eagle library.
This custom node is a ComfyUI node for generating speech from text using the Gemini 2.5 Flash Preview TTS API.
ComfyUI custom nodes for LM Studio integration with conversation history support
Smart resize node for ComfyUI that handles portrait/landscape images with short/long side specification
ComfyUI custom node for removing specified tags and their content from text
NODES: Upload image(s) to WebDAV, Upload video as WebM to WebDAV, Load image from WebDAV, Evaluate S-Expr with integer output, Evaluate S-Expr with float output
A collection of custom nodes for ComfyUI focused on image processing operations.
These nodes allow you to make Checkbin comparisons.
A growing collection of practical, high-quality nodes designed to extend ComfyUI with useful utilities, smarter workflow controls, and flexible generation tools.
A node to replicate https://huggingface.co/spaces/latentexplorers/latentnavigation-flux, the initial node is made by https://github.com/RhizoNymph/ComfyUI-CLIPSli der created, I made a few changes to make it more in line with my needs.
Whole-clip sampling preview for MiniMax H3, decoded on the CPU with taeh3 so it costs no VRAM.
NODES: SiliconCloudReasoning, DeepSeekOnline
add http api http://127.0.0.1:8188/comfyui-run/run use in other llm project.
ComfyUI: Novakid. A node.
Custom node that stores 10 presets, each bundling HIGH/LOW LoRA pairs and prompts, for efficient workflow switching by single integer input.
Bidirectional tile-centric mixed-precision (FP16/INT8) attention for ComfyUI diffusion transformers, with a Krea 2 gated-MMDiT patch.
Structured + chaos-based prompt generation system for ComfyUI. Generate less. Evolve more.
NotifyMe sends task IDs and optional text to an allowlisted HTTPS-only callback with any-type passthrough. English node UI. Also includes independent PNG media carrier, metadata-filtered saving, and PNG image restoration nodes.
ComfyUI custom nodes wrapping Google Gemini APIs for image generation, multi-image composition, video generation (Veo and Gemini Omni Flash), and local LaMa inpainting.
NODES:Frame Segmenter, Get Frame at Index, Repeat Sampler Config, Patch Repeat Sampler Config (Model), Patch Repeat Sampler Config (Latent), KSampler (Simple Input)
ComfyUI custom node for Semi-supervised Parametric Real-world Image Harmonization (PIH, CVPR 2023)
A collection of useful user interface extensions for ComfyUI to improve workflow and provide helpful feedback.
Resizable DOM category text fields that assemble an H3 brief for Amp NInfer.
A collection of quick-use custom nodes for ComfyUI - image processing and workflow tools
ComfyUI node: Get DateTime (outputs date, time, datetime as strings
Random selection nodes with freeze/lock capability for iterative prompt building in ComfyUI
A ComfyUI custom node that calculates semantic similarity between two images using CLIP Vision models, outputting a numerical similarity score. (Description by CC)
Custom node to detect an edit in a batch of images.
Image generation nodes powered by OpenAI's GPT Image 2 models, supporting text-to-image and image-to-image transformations with configurable parameters.
Custom node to extract the colors in an image as a palette for use with ComfyUI-PixelArt-Detector
This is a set of custom nodes for ComfyUI, designed for the following application: [a/https://github.com/cnnmmd/cnnmmd](https://github.com/cnnmmd/cnnmmd)
ComfyUI nodes for OpenRouter video, dialogue and music, with cast and frame stacks, a LoRA browser, and a latent chooser.
Wire Loom nodes for organizing and bundling wires in ComfyUI workflows into reusable streams. (Description by CC)
Custom ComfyUI nodes that verify required files/models are present per workflow and download them if missing.
Desktop LoRA recipes, automatic triggers, FRAME prompt libraries and switch-only phone workflows
A custom node for ComfyUI that provides advanced, condition-based prompt appending allowing you to dynamically build prompts based on the presence or absence of specific tags.
Chat with an LLM in the ComfyUI sidebar, write image prompts with st-chatu8 style presets, then fill a node and queue the workflow in one click.
Local prompt browsing, composition and editable snapshots for ComfyUI
Quickly open the source code to nodes in ComfyUI.
A ComfyUI custom node that runs N×M prompt-variation grids through the full sampling pipeline in a single node with labeled output, flexible axes, and optional node batching.
This node automatically pastes copied images and text.
ComfyUI nodes by ComfyUIdev. Favorite jumps to a saved canvas anchor. Workflow Helper is a shareable workflow card with a Markdown how-to, model downloads, and pack checks.
A custom node that use for donwload models, fork from @Kosinkadink
Nodes for simple frame interpolation without the use of AI. Uses standard image operations to blend frames together.
Node comparing two image masks or images with adjustable overlap threshold (default 95%) for detecting minor shifts and mismatches in proportions, suitable for automated post-processing validation. (Description by CC)
GPT Image sizing and automatic edit preparation/restoration with original-image geometry and mask compositing.
collection of nodes for prompt batching and management
A ComfyUI custom node for parsing CSV files with positive and negative prompts
Check one image for IP similarity using CopySight. Returns an annotated image and JSON report. Requires an enabled CopySight key and service credits.
Generate SMPL / SMPL-H body avatars from height and weight in ComfyUI: shaded render, exact silhouette mask, depth map, 24 body-part masks and OpenPose keypoints, from one view or a set of views (character sheets, orthographic or perspective), for ControlNet-style conditioning.
MengBaoAI | 萌宝AI ComfyUI 综合节点套件:图像生成、图片复刻、电商设置、提示词库、素材库、历史记录和图像处理。Image generation, replication, e-commerce and image tools.
A power-user focused collection of ComfyUI nodes: LoRA management, prompt building, resolution presets, metadata-safe image saving (CivitAI-ready), and UNet model loading.
Create transparent PNG images from RGB + mask input with automatic bounding box cropping. Supports mask inversion and flexible image/mask size mismatch handling.
Google or offline CPU translation for ComfyUI with preserved quotes, backticks, and inline previews.
A tiny ComfyUI custom node that saves PNG or JPEG images without ComfyUI prompt/workflow metadata.
This is a subset of nodes for ComfyUI that I made just for my own workflow. The nodes support Flux (single conditioning, no negatives) and are just a way of minimizing the noodles. There's no real journey to be made here, just anything that helps me get through the day.
Utility node for ComfyUI that calculates downscaled resolutions while maintaining original aspect ratio, ideal for image-to-video workflows to prevent OOM errors. (Description by CC)
Enhance, denoise, restore and retouch images with CraterView.ai. Each node sends an image to the API and returns the result, with no graphics card or model files.
generate tags / prompt from danboru image_id input
generate tags / prompt from danboru image_id input
A set of custom nodes for ComfyUI that merge video and audio using ffmpeg.
A Gaussian blur node for ComfyUI using OpenCV
A node that can use Nai in Comfyui
Model-independent hex palette construction and perceptual image harmonization for ComfyUI.
HeartMuLa music generation + lyrics transcription with MuQ-MuLan reference-audio style transfer (fork of ComfyUI-HeartMuLa)
Persistent memory and multi-agent communication for ComfyUI — powered by AgentAZAll
Nodes for identifying target regions in images, extracting UV vertices, and writing metadata with preview generation. (Description by CC)
A focused set of CubicJ custom nodes for ComfyUI.
Generate video, images, speech and music in ComfyUI with your Cuddler Platform API key. Prices are shown before you run, and every run has a spending limit.
CLAHE preprocessing node for improving 3D mesh generation with TRELLIS
explainable AI for comfy
ComfyUI custom node for Freepik Seedream v4.5 image generation and image editing.
A workflow-portable prompt library node for ComfyUI with remote catalog syncing and image-first prompt browsing.
ComfyUI custom node for converting IMAGE batches into copyable PNG or MP4 Base64 strings, with optional MP4 audio muxing.
Standalone CyberKrea Enhancer Advanced node for ComfyUI Krea2 models
Preset sigma schedules for Z-Image Turbo and Anima-Turbo in ComfyUI
Two regional characters with their own LoRAs in one Krea 2 image, with optional NegPiP negatives.
Prompt cleanup for ComfyUI — port of sd-webui-prompt-format. Dedupe tags, fix bracket/comma spacing, swap underscores. Button-driven, not realtime.
Two Z-Image Turbo characters with independent regional LoRAs, prompts, free boxes and masks in one ComfyUI node.
A minimal template for creating React/TypeScript frontend extensions for ComfyUI, with complete boilerplate setup including internationalization and unit testing.
A ComfyUI extension providing optimized resolution presets for multiple AI image generation models with support for various aspect ratios and two node variants.
ComfyUI custom nodes for project management.
ComfyUI custom nodes for general purpose.
Tensor manipulation in ComfyUi
Nodes:FaceCompare
Prompt warehouse and workflow utility nodes for ComfyUI, including reusable prompts, multiline prompt editing, a persistent multi-LoRA loader, and deletable image output.
This node generates a professional prompt from an input text for modern video AI models (e.g., Alibaba Wan 2.2) via the OpenRouter API.
Two ComfyUI nodes: a picture loader you put in order yourself, and a shot node for size and clip length.
A node to drop frames from the output images to force the video to a lower FPS at the same playback speed.
Smart real-time token counter for CLIP-L, T5-XXL (Flux), Qwen + seed-controlled prompt mixer from custom CSV wildcards. Supports field overrides, flat/paragraph modes, thread-safe caching.
One node generates both T2I & I2V positive/negative conditioning in a single click.
Silently keeps the DaSiWa ComfyUI workflows (darksidewalker/dasiwa-comfyui-workflows) up to date in your Workflows menu, with an Active-only / All setting.
LM Studio prompt writer for ComfyUI (Krea2 / Illustrious) with reference images, per-image toggles and an aspect/MP size node.
ComfyUI helper node that converts JSON rectangle regions into SAM3_BOXES_PROMPT for ComfyUI-SAM3.
Speech bubble detector for ComfyUI based on SAM3 Text Segmentation with crop-locked region detection for manga/manhwa cleanup workflows.
Reproduce torch 2.8.0+cu129 CUDA numerics on newer torch, so MiniMax Music 3 / YuE2 seeds give the same songs as before the torch upgrade.