ComfyUI Custom Extensions
Browse 9,446 ComfyUI extensions and 73,215 custom nodes.
a simple white extractor node for comfyui.
Miscellaneous utility nodes
A simple scaling node for ComfyUI, allowing the changing of the length of the input's longest edge.
Get a tag-based prompt from a booru post.
A collection of Logic Nodes that I think are missing in core ComfyUI.
A collection of Logic Nodes that I think are missing in core ComfyUI.
ComfyUI nodes: offline Chinese speech-to-text (sherpa-onnx + Paraformer-zh-small, pure CPU, no network, no VRAM) + payment-amount extraction. 离线中文语音转文字 + 收款金额抽取。
Take the last N frames of an IMAGE batch regardless of total frames (handy for multi-modal video segment stitching) plus a compat shim that re-registers the lost MySplitImageBatch node.
ComfyUI node: stitch multiple video segments (with audio) seamlessly — streaming, constant memory, no OOM. 视频无缝拼接,防 OOM。
ComfyUI custom nodes for high-level image generation and editing with Vector Engine Gemini API integration. (Description by CC)
ComfyUI custom node for grouped anime illustration style prompt injection.
Integrated beam-search-only fork of pharmapsychotic's ComfyUI-Cliption, combining the CLIPtion loader and beam search into a single node
Text to 3D and Image to 3D in ComfyUI. Create textured GLB assets with your Cormak account, preview them and use them in 3D workflows.
ComfyUI custom nodes for depth-to-layers processing and grid slicing. For Game Development use cases.
Hold MiniMax H3 extends made with the obvpm Timeline to the tone of the clip they continue, and check how far they moved
A ComfyUI custom node that removes spikes from masks and converts them to smooth elliptical masks with feathered edges.
FFT-based image preprocessing nodes for ComfyUI, designed to reduce latent instability in Flux image-to-image workflows.
A collection of custom nodes for ComfyUI
Universal Regional Prompter with a paintable region map, Universal Model Loader, and optional-input checkers for ComfyUI.
Experimental Krea-2-specific sampling nodes for ComfyUI with advanced solver using Adams-Bashforth-2 history correction and curvature-aware flow integration. (Description by CC)
Auto Switch (ANY) by Steve Lasmin — A ComfyUI custom node for automatic signal routing between two universal inputs.
A ComfyUI custom node that loads images sequentially from a folder, cycling back to the first image when reaching the end. Outputs the loaded image along with its width and height as strings.
Calculates width and height from max resolution, aspect ratio, and multiplier for ComfyUI
A ComfyUI routing node that auto-selects exactly one active signal from 5 optional ANY inputs and routes it to the output. Raises a red-node error if zero or multiple signals are detected.
All-in-one ComfyUI workflow hub: resolution calculator, noise, sigmas, conditioning, and sampler outputs
This repository contains a custom ComfyUI node for overlaying media using ffmpeg.
Toggle-Pass: Privacy toggle, dynamic LoRA slots, and other quality-of-life nodes for ComfyUI
Preview native ComfyUI VIDEO output without permanently saving every generation.
Custom nodes for ComfyUI that split multiline text prompts into multiple independent STRING outputs, with each line becoming its own text output ready to connect to CLIP Text Encode.
A simple wildcard prompt node with seed control and debug output.
This extension provides object segmentation capabilities for ComfyUI workflows
Test Custom Node - Sample Node for testing the publish-node-action
A ComfyUI custom node for publishing images to MQTT brokers with support for TLS encryption and authentication.
Custom ComfyUI nodes for fal.ai models
Concat conditions and prompts for ComfyUI
ComfyUI nodes for Qwen-Scope SAE interpretability and steering.
Embeddr ComfyUI Extension
A growing collection of EmberFrame utility nodes for Comfy, including advanced PiD sampling, Z-Image/Flux latent normalization, wildcard prompt helpers, and resolution tools.
Research-grade CLIP Score metric for Flux/SD. Raw cosine similarity.
Nodes:MultiTextEncode, KolorsMultiTextEncode, Caption, BatchImageProcessor
Additional custom samplers, sampling nodes, and CFG guiders for ComfyUI.
ComfyUI nodes for local Qwen3-ASR transcription.
A ComfyUI plugin for integrating with SGLang Diffusion server, supporting image and video generation capabilities.
Lazy-loading ComfyUI fork of Ultimate Stable Diffusion Upscale.
Workspace for a new high-quality, high-performance hires-fix / upscale custom node for ComfyUI.
Version 1.2.1 - Dependency cleanup and archived LineSelector node A comprehensive collection of custom nodes for ComfyUI, providing advanced image processing, workflow control, and utility functions to enhance your AI image generation workflows.
Lean ComfyUI custom nodes for in-memory VACE clip collection, seam prep, and iterative joins.
ComfyUI custom node pack for scouting and rendering calibrated camera shots over Gaussian splat scenes.
Image-editing drift correction and experimental Qwen Image 2.1 text encoding for ComfyUI.
Shift-click color picking for ComfyUI's mask editor Paint Pen
Custom utility nodes for ComfyUI
ComfyUI custom node that loads images (including HEIC/HEIF) and provides browser-friendly previews.
ComfyUI custom node: load a LoRA from current selection and automatically advance current within a first/last filename range.
A single, no-nonsense node to pick text-to-image dimensions from an aspect-ratio dropdown, a long-edge slider, and an invert toggle. Outputs width/height (INT) plus an info STRING that previews the resolved size, orientation and common long-edge reference sizes.
Display prompt-ready Picture, Video, and Audio labels on ComfyUI's MiniMax H3 reference node.
ComfyUI custom nodes for STAR, a diffusion-based 4x video upscaling model with temporal coherence, VRAM offloading, and long-video support. (Description by CC)
Tiled upscaling for ComfyUI using built-in nodes. Replicates the core features of UltimateSDUpscale as a transparent workflow you can inspect and modify.
A custom node for ComfyUI to load and run Qwen3 GGUF models. This node is designed to work with AMD GPUs via ROCm, as well as CPU backends.
A node for saving cleaned text outputs, useful for LoRA training. Removes unwanted tokens like <pad> and saves to .txt.
Browse Civitai inside ComfyUI: a discovery feed, your favorites/collections and the artists you follow in the sidebar, with tag and text search, base model filters, one-click prompt/seed/image transfer into your workflow, reactions and collection management.
ComfyUI custom nodes for MOSS-TTS 1.7B zero-shot voice cloning
A collection of custom nodes for ComfyUI
Retired: Phantom Router no longer accepts workflows from ComfyUI. Remove this node.
Official EvoLink nodes — 56 nodes, one node per model variant: Nano Banana 2/Pro/Lite, Seedream 5.0, GPT Image 2, Qwen Image 3.0, Z-Image Turbo, Krea 2, Grok Imagine, Sora 2/Pro, Veo 3.1, Kling V3/Turbo/O3 (ref/edit/motion), Seedance 2.5/2.0 (ref/edit/extend), Wan 3.0, Hailuo 2.3, Gemini Omni Flash, Topaz Upscale. Chain generate→edit→upscale via result_urls. No local GPU, no third-party image hosting, async & cancellable, API key stays out of workflow JSON. 国内直连,中文支持。
Spectrum-style acceleration path for FLUX in ComfyUI.
ExoticArts custom nodes for ComfyUI (EA Power LoRA, etc.)
ComfyUI custom nodes for the Labnana (Marswave) image generation API: one node covers text-to-image, image editing and 4K output, plus credit estimation, task history and subscription tools.
ComfyUI model manager for Stability Matrix shared folders, built to work alongside LoRA Manager: missing-model downloads, CivitAI library, base-model organizer, recipes from images, dedupe, metadata sync and SSD/HDD storage tiers.
Retired (no longer developed; see ComfyUI-Model-Librarian). ComfyUI side of the Prompto Chrome extension: Prompto Recipe, Character, Snippet, Wildcards, Profile and LoRAs nodes, a Prompto library sidebar, load recipes into the workflow, add snippets at the cursor, Save to Prompto on nodes and images, and auto-capture of saved images.
Translates and enhances prompts from multiple languages to English using local GGUF models.
Cinematic video prompt tools for ComfyUI — ThreeShot sequences, storyboard assembly, scene camera moves, and vision-driven parameter extraction via CLIP, Qwen, Gemini, ChatGPT, or Claude.
A custom ComfyUI node that displays live processing time in a red digital countdown clock format, perfect for monitoring image generation times and tracking performance between workflow nodes.
Comfyui Custom Nodes for Showrunner
A custom node for ComfyUI that lets you visually set up a lighting scene and automatically generate a detailed prompt string describing it.
Adjust the color of image before hires.fix
Extension that allows to use emphasize and AND on Z-Image.
Just a small and random collection of nodes for testing and finer control over the diffusion process, so it's actually useful as a tool in my workflows rathen than just a slop machine.
A custom ComfyUI node for extracting modern, high-end architectural color tones and atmospheric keywords from reference images. (Description by CC)
ComfyUI custom node wrapping Meta AI's Eulerian Motion Fields research model for fluid cinemagraphs.
Attempt to create a ComfyUI node for masking AI-generated fingerprints.
Integrates Freepik's AI capabilities into ComfyUI workflows with features for photorealistic generation, upscaling, and background removal, plus smart caching and cost management. (Description by CC)
Houdini-style procedural geometry (SOPs) and Copernicus-style image nodes (COPs) for ComfyUI: 500+ numpy nodes with Houdini-style parameter panels, a 3D viewport with handles, expressions, keyframes, For-Each loops and a geometry <-> AI texture round trip.
Unofficial GGUF companion package for JoyAI-Image-Edit, adding GGUF loader nodes as drop-in replacements for bf16 transformer and text encoder loaders.
ComfyUI plugin for face detection, square cropping with dilation, and paste-back with feathered edges for smooth blending.
A suite of powerful and versatile utility nodes for ComfyUI, designed to streamline complex workflows involving Large Language Models and text manipulation.
This is a ComfyUI plugin based on [a/Spleeter](https://github.com/deezer/spleete). The model files are available on BaiduNetDisk. After downloading the model, place it in the checkpoints directory.
Independent schema-aware CRUN models, Effects, AI Tools and safe starter workflows for ComfyUI.
ComfyUI custom nodes for calling Pi as a text and optional image LLM from a workflow, with local caching for reproducible results and companion text extraction utilities.
Maya-style global timeline for ComfyUI — a shared bottom-panel bar (scrub, range, transport, keyframe dope-sheet) + curve editor + frame-accurate viewer, with a public JS & Python API any node can connect to.
Self-contained ComfyUI remove core extension with region editing and mask processing for image inpainting. (Description by CC)
A fast tool utility for ComfyUI
This repository contains custom nodes for ComfyUI, designed to enhance your workflow with text-based operations, particularly for managing and utilizing descriptive texts.
Comfy Custom Node for Try-on.
ComfyUI nodes for nobg: open-source background removal and image matting.
Repository containing two ComfyUI nodes for offline translation using translateLocally, enabling text translation and model selection without internet connection.
Bernini-R omni video generation nodes for ComfyUI with modality-aware rendering, model loading, progress logs, and image/video outputs.
Photoroom PRXPixel text-to-image nodes for ComfyUI with live previews and patch-grid controls
Fill Labs VLM nodes for ComfyUI, starting with Qwen3-VL image and image-batch inference.
Prompt handling and workflow state control nodes for ComfyUI.