ComfyUI Custom Extensions
Browse 9,446 ComfyUI extensions and 73,215 custom nodes.
Fish Audio Dify-parity tools through Ace Data Cloud for ComfyUI
Flux Dify-parity tools through Ace Data Cloud for ComfyUI
Google Search Dify-parity tools through Ace Data Cloud for ComfyUI
GPT Image Dify-parity tools through Ace Data Cloud for ComfyUI
Grok Dify-parity tools through Ace Data Cloud for ComfyUI
Happy Horse video generation, editing, task lookup, and native video loading through Ace Data Cloud
Kling Dify-parity tools through Ace Data Cloud for ComfyUI
Midjourney Dify-parity tools through Ace Data Cloud for ComfyUI
MiniMax H3 Dify-parity tools through Ace Data Cloud for ComfyUI
Nano Banana generation, editing, task lookup, and image loading through Ace Data Cloud
Qwen Image Dify-parity tools through Ace Data Cloud for ComfyUI
Seedance Dify-parity tools through Ace Data Cloud for ComfyUI
Seedream Dify-parity tools through Ace Data Cloud for ComfyUI
Generate Suno music through your Ace Data Cloud account, query the same task ID, and load completed audio into ComfyUI. Requires an Ace Data Cloud API key; generation consumes Credits. Cloud API only, with no local model or GPU needed.
Veo generation, editing, task lookup, and native video loading through Ace Data Cloud
Wan Dify-parity tools through Ace Data Cloud for ComfyUI
A collection of ComfyUI custom nodes and utilities by acgc99.
Custom nodes for ComfyUI that generate multiple images in a single GPU batch with different seeds.
Custom ComfyUI node that detects and extracts a single loop cycle from a batch of images. (Description by CC)
Custom ComfyUI node for generating prompts by combining manual character descriptions, outfit sections, and .txt file lines. (Description by CC)
ComfyUI custom node for OpenRouter chat completions with vision input, reasoning effort selection, and JSON extraction. (Description by CC)
Custom ComfyUI nodes for GROUT (Geometric Reasoning Over Unstructured Tessellations) - a deep learning model for detecting and segmenting grout lines in mosaic images.
This repository contains a powerful and versatile custom node for ComfyUI that seamlessly integrates with OpenAI-compatible Large Language Models (LLMs), including multimodal (vision-enabled) models like GPT-4o. This single node allows you to perform both text generation and image analysis, making it an essential tool for advanced prompt engineering and creative automation.
ComfyUI nodes for OpenAI gpt-image generation and editing
Solves the complex 'For-Loop' logic in ComfyUI. Perfect for controlling multi-subject vs. multi-scene batch workflows.
Generic mesh manipulation nodes for ComfyUI. Works with any 3D mesh — not tied to any specific generation pipeline.
ComfyUI nodes for ByteDance Seedance 2.0 video generation via the MyteamOne API — text/image/video to video, plus multi-modal reference.
Extract individual actors from video with green screen background, face-based identity clustering, and structured metadata output
Video file management and merging capabilities, fully compatible with the VideoHelperSuite plugin ecosystem.
Split a prompt into named elements and save the whole set as a preset. Weighted random syntax, labelled presets, tag autocomplete.
Generate images on remote GPUs from ComfyUI with Badgr, with a hard cost cap per run.
A custom node for ComfyUI designed to streamline the prompting process when using the FreeFuse workflow.
Small, dependency-free utility nodes used by AICU's ComfyLTS lines and books. Starts with Resize Shortest Side To.
Local NVIDIA Maxine Eye Contact for ComfyUI VIDEO with progress and experimental iris-colour preservation. Requires a separately configured NVIDIA NIM service.
a custom node for [a/DiffSynth-Studio](https://github.com/modelscope/DiffSynth-Studio)
ComfyUI node for fast-whisper transcription
A dead-simple ComfyUI node that resizes image(s) to a chosen megapixel count, preserving aspect ratio, with optional dimension rounding for diffusion models.
ComfyUI custom node that generates shareable HTML UIs from ComfyUI API workflow JSON.
Timeline builders for the MiniMax H3 Material Planner and the LTX 2.5 Director, a Gemma 4 attention fix for ROCm, and a chunked ESRGAN video upscaler.
Windows-friendly edition of the AnimaForge ComfyUI Anima LoRA factory with Windows-first layout and setup. (Description by CC)
Hubris nodes for ComfyUI: hundreds of text models, Nano Banana, GPT Image, FLUX, Seedream and Recraft for images, Kling, Veo, Seedance, Wan and Hailuo for video, and text to speech, all with one API key.
Nodes:Merge Tag, Clear Tag, Add Tag, Load Images Pair Batch, Save Images Pair
LoRA routing, prompt and node history, canvas packing, and scene handoff for ComfyUI.
ComfyUI nodes for iterating over files (images, text) in directories.
Fills in missing models through a configurable mirror (hf-mirror by default): detects missing model files in the current workflow, resolves each to its official HuggingFace repo by filename, and downloads them into the correct models/ folder with resume support. Writes only inside models/ (validated allowlist, no path traversal). Does not patch any official ComfyUI file.
Custom nodes for ComfyUI — Ratio Outpaint Calc and BBox Multiple Fix
Custom ComfyUI nodes for calling Seedance 2.0 through a direct BytePlus ModelArk API key.
A ComfyUI node for the CSM model featuring text-to-speech, voice cloning, and automatic model downloading from Hugging Face.
Custom ComfyUI node that implements the Guided Meta-Sampler for video outpainting.
Save JPG and PNG images with their original filenames and image previews in ComfyUI.
Save VideoHelperSuite MP4 outputs with source filenames, optional intermediate cleanup, and automatically calculate RIFE interpolation FPS.
Generate images and videos in ComfyUI with the Sakuranbo AI API.
Mark AI-generated images for the EU AI Act (Article 50): invisible watermark + signed C2PA Content Credentials, saved straight from your workflow. Plus a watermark verify node.
ComfyUI nodes for our product Faishme
A clean and simple ComfyUI custom node that creates a LoRA Hook and automatically finds and loads trigger TXT files placed in a folder named after the LoRA file.
Lineage tracking, visual clustering, and a web UI for managing generated images in ComfyUI.
Misc Text Nodes for Comfy UI
A drop-in replacement for EmptyLatentImage in AnimateDiff and video workflows, controlled entirely by dropdowns.
One node to drive resolution, clip length, and frame rate across a ComfyUI video graph from a single control.
A ComfyUI node for extracting perspective views from 360 degree equirectangular images with an interactive drag-and-drop crop interface.
Custom ComfyUI nodes for Beeble SwitchX image and video workflows.
AlcAI Nodes is a collection of custom nodes for ComfyUI, a modular interface for image generation with Stable Diffusion. This project focuses on tools optimized for creating anime-style images and stylized effects, inspired by the AnimeHaven universe. It aims to simplify workflows for artists and AI creativity enthusiasts.
Call external APIs: remote ComfyUI, Grok, OpenAI-compatible inference, zero-shot classification [a/Inference](https://github.com/alchemine/comfyui-api-pack/blob/main/workflows/comfyui-api-pack-workflow-Inference.png): OpenAI-compatible API Inference [a/ZeroShot](https://github.com/alchemine/comfyui-api-pack/blob/main/workflows/comfyui-api-pack-workflow-ZeroShot.png): zero-shot classification [a/API](https://github.com/alchemine/comfyui-api-pack/blob/main/workflows/comfyui-api-pack-workflow-API.png): remote ComfyUI API execution
Danbooru tag retrievers and posts downloader
Image enhance pack. - Adjust Image(color correction, sharpen, denoise, resize), Despeckle Image - [a/Workflow](https://github.com/alchemine/comfyui-enhance-pack/blob/main/workflows/comfyui-enhance-pack-workflow.png)
Image evaluation pack. - Anatomy Evaluator: scores faces, eyes, hands, fingers and feet for count and sharpness, with a report of every check
A collection of custom image processing nodes for ComfyUI
Save checkpoint .safetensors with custom header JSON, optional prompt override, merge of EXTRA_PNGINFO, and smart/no-counter filename modes. Includes text outputs for path and final metadata.
A modular collection of custom nodes for ComfyUI, organized by data type for easy navigation and maintenance.
Generates rectangular mask from coordinates with optional padding.
Simple box padding for Comfyui [Simple Padding](https://github.com/alexlawford/comfyui-simple-padding).
ComfyUI V3 node that uses Claude to optimize prompts for current image and video generation models.
Configurable output path generator with dynamic segments and live preview for ComfyUI.
Sends ComfyUI workflow provenance (outputs, references, prompts, models) to Archibal.
A ComfyUI extension that adds a sidebar for browsing LoRAs with preview images and descriptions
A ComfyUI extension that adds a notes sidebar for managing notes
Client-side Russian reference for ComfyUI nodes
A ComfyUI custom node that allows you to pipe a generated prompt into from an LLM and either pass it as is, or copy and edit it manually.
ComfyUI model metadata editing nodes.
Resolution Picker and Calculator
A ComfyUI custom node that scans PyTorch model files for potential security vulnerabilities without executing them.
Comfy Flow - A flow manager for ComfyUI. (Description by CC)
Henge nodes for MiniMax H3 video editing: shot loading, 2D/3D color guides and latent masks from SAM 3.1 tracking (scene by scene across camera cuts), H3 guide-latent conditioning with optional CLIP video context, people finder, subject looks, causal sampler, frame refine, H3 latent / NVIDIA RTX upscalers and video save. Used by the Henge studio (github.com/alisson-anjos/Henge).
In a dual sampler workflow with mixture of experts setup where one model provides the early steps and another provides the late steps, it may be useful to switch the order of the models.
Latent detection logic node
Place image stickers on the surfaces in a photograph using MoGe geometry.
This plugin adds a new node to ComfyUI. The node uses the FLUX.1 Fill API to fill in parts of an image.
A custom ComfyUI node that provides an interactive video player with timeline scrubbing and audio waveform display.
Seamless-loop / FLV (first-last video) drift compensation: registers every frame of an image batch against one reference framing derived from the first/last frames (ORB + RANSAC affine/similarity), with optional radial-distortion correction, Lie-algebra log interpolation, and first/last loop blending.
Seamless-loop / FLV (first-last video) drift compensation: registers every frame of an image batch against one reference framing derived from the first/last frames (ORB + RANSAC affine/similarity), with optional radial-distortion correction, Lie-algebra log interpolation, and first/last loop blending.
Centralized video timing, FPS, resolution, latent preparation and video metadata for ComfyUI workflows.
Audio file upload, playback, waveform selection and duration-based AUDIO output for MiniMax H3 reference audio.
Video timing, fixed native H3 FPS, resolution presets, automatic size alignment and shared metadata for MiniMax H3 workflows.
An advanced Empty Latent selector with visual presets and foldable UI.
ComfyUI nodes for SeFi-Image (Semantic-First Diffusion) text-to-image.
Nodes: RErouter, String (RE), Int (RE)
Nodes for removing UI elements from screenshots by detecting and masking icons, buttons, and text. (Description by CC)
Custom nodes for ComfyUI providing image loading and canvas manipulation capabilities.
A ComfyUI custom node project that extracts sonic features from audio and uses a local Ollama LLM to generate painterly, emotionally-driven image-generation prompts.
Valida se a imagem é múltipla de 8 para workflows de Upscale e Flux.