ComfyUI Custom Extensions

save images to SCP upload, direct to webserver

A ComfyUI custom node pack that provides text-free, deterministic CLIP conditioning for Stable Diffusion image generation using integer seeds instead of text prompts.

zerobytes procedural EDIT system for instruction based image generation with custom word pools

A ComfyUI custom node that draws a status bar over the top of each image in an input batch.

Qwen25-VL is a plugin for ComfyU

Pass string and images between multiple ComfyUI via socket. Easy and intuitive nodes for remote connection.

Custom node to generate a image using NanoGPT API

Save images as PNG/JPEG/TIFF/PSD with DPI metadata and filename tokens (%model%, %seed%)

Text and data manipulation nodes for ComfyUI, with emphasis on JSON processing, detection workflows, and bbox visualization.

Custom nodes with split, random, and select functions for easy visual and management of multiple cumbersome prompts

Wavelet-based frequency-aware noise coloring for ComfyUI samplers; adapts noise injection across frequency bands during generation to improve image detail and quality. (Description by CC)

Nanaix image generation and editing nodes for ComfyUI

A collection of logic and utility nodes for ComfyUI to enhance workflow capabilities.

Custom node set for ComfyUI to assist in constructing prompts for Anima text-to-image generation with structured assembly, Danbooru tag validation, and tag normalization. (Description by CC)

Local-only output nodes that derive a stable run key, save image hashes, and write a receipt beside ComfyUI images. Repeated identical outputs are reported without overwriting the completed receipt.

PuLID for ComfyUI with Chroma support with changed class names to avoid conflicts

A custom node for ComfyUI that adds text overlay to images, with options for text and background color, opacity, and vertical positioning.

ComfyUI custom nodes for T5Gemma-TTS, a multilingual text-to-speech model with voice cloning and duration control, based on the T5Gemma encoder-decoder LLM architecture.

neggles custom comfy nodes.

This node pack provides a Python node that uses the SambaNova API to send prompts to a chat AI model (e.g., DeepSeek-V3-0324) and retrieve responses, intended for integration into node-based workflows like ComfyUI.

A ComfyUI custom node that blends two images in latent space using a mix factor slider.

NODES: Better Image Dimensions, Standard SDXL Dimensions node, Dimensions by Ratio node, ...

Scales the height of an image to the closest number divisible by 16, based on an input width and the aspect ratio of the image.

Automatically generate captions for audio files.

Since I used 'simple wildcards' from Vanilla and it no longer works with the new Comfy UI version for me, I created an alternative. This CustomNode takes the entire contents of your wildcards-folder(comfyui wildcards) and creates a node for each one.

Universal Replicate API integration for ComfyUI - video (Sora 2, Veo 3.1, Seedance, Kling), image (Nano Banana Pro/2, FLUX), lipsync, audio and 3D models with multi-image reference input. 通用 Replicate API 整合，支援影片/圖片/音訊/3D 生成與唇語同步。

A ComfyUI extension for generating captions for your images. Runs on your own system, no external services used, no filter.

A ComfyUI custom node that generates/refines prompts using a local Qwen3-VL/llama.cpp backend with editable system prompts and per-preset max_tokens. (Description by CC)

A dependency-free ComfyUI custom node that expands combinatorial prompt syntax.

A ComfyUI custom node that detects NSFW content using Freepik/nsfw_image_detector (EVA-based, ~86MB) and saves images with detection results in the filename.

A lightweight ComfyUI extension designed to automate the creation of video contact sheets (preview grids).

LiveLatentPreviewer & Saver (LLPS) provides live preview and disk saving functionality with per-node control via LLPS KSampler, including configurable preview methods and save intervals.

Lets you open a generated Image in a specified application after generation.

A lightweight custom node for ComfyUI that sends an HTTPS GET request upon the completion of a workflow.

Adds a debug viewer on the right panel

MiniT2i for ComfyUI

Video FaceDetailer for SDXL in ComfyUI: face tracking, per-track fixed-seed detailing and flow-guided temporal blending to fix faces in videos without flicker.

ComfyUI plugin for text-to-image and image-to-image generation using Seedream 4.5 model from Volcano Engine. (Description by CC)

MapRoute concept erasing method for ComfyUI (CVPR 2026). Load trained Mapper weights to erase specified concepts from generated images. (Description by CC)

Custom Image processing ComfyUI Nodes

An experimental ComfyUI node pack containing a vision loader to load the SigLIP vision encoder from Gemma3 and a text conditioning node for image-text integration. (Description by CC)

Essential little utilities that should have been in ComfyUI from day one. (Description by CC)

This is the transformer-based implementation of Niutonian GLM-4.6V nodes for ComfyUI with memory optimizations.

A comprehensive ComfyUI custom node suite for intelligent image stitching, processing, and manipulation with automatic handling of empty/bypassed inputs.

ComfyUI custom nodes that sample the edge colors of generated images and send matching colors to Philips Hue lights or Hue light strips.

A custom node for ComfyUI that allows you to easily iterate through all of your downloaded checkpoints sequentially or dynamically.

This node is for playing the game of guessing prompts by looking at images generated from prompts output by TIPO, Tagger, etc..

A suite of ComfyUI custom nodes based on the RWKV series of models, offering various functions like translation and dialogue.

Easy save image with dir+name

顔→人物生成でブレる頭サイズ比(頭幅/肩幅)と顔の一致を、参照に近い順へ自動ランキング。顔類似×頭サイズ比×鮮鋭度の合成スコアで選別し、合格分だけ納品するComfyUIカスタムノード。

Umbra Studio custom nodes for ComfyUI metadata saving, Power Prompter integration, and workflow bridge support.

A lightweight ComfyUI extension that adds a floating RGBA color picker panel for coloring selected nodes on the canvas.

Custom ComfyUI nodes by NOLABEL for studio workflows, featuring Shot Path Builder that generates standardized, sanitized file and folder names with versioning for renders.

Short and simple custom node for comfyui Manhwa processing.

Provides Qwen Image Edit custom nodes: Text Encode with reorderable text slots for prompt building and Skin Realism post-processor. (Description by CC)

A single ComfyUI custom node that computes latent-safe width/height from an image's aspect ratio and can output a matching empty LATENT batch.

A lightweight ComfyUI custom node for blending two image batches using a per-frame scalar mask without requiring full resolution mask upscaling.

Pack custom node SDXL untuk ComfyUI dengan arsitektur modular (bukan single script), supaya lebih mudah di-maintain, di-scale, dan aman untuk workflow production.

Gemini vision integrations into ComfyUI.

A simple ComfyUI custom node to enable Kernel Samepage Merging (KSM) on Linux, allowing memory deduplication to reduce RAM usage depending on workload.

Upload the image to Alibaba Cloud OSS.

使用阿里云oss上的图片,或者保存结果图片至阿里云oss

Custom ComfyUI nodes providing video comparison utilities, including wipe and fade compare video generation in multiple directions. (Description by CC)

Photoroom PRXPixel (prxpixel-t2i) — pixel-space text-to-image (~7B DiT + Qwen3-VL) for ComfyUI.

ComfyUI nodes for Qwen-Scope SAE interpretability and steering.

Utilities to manage zips and directories

Anime character image evaluation nodes (CCIP + OKS + Angle) for ComfyUI

ComfyUI custom nodes for human OK/NG judging of generated images, oriented toward LoRA dataset curation.

Nodes:Image Crop Fit, OFF SEGS to Image, Crop Center wigh SEGS, Watermarking, GW Number Formatting Node.

The Intelligent AI Assistant for ComfyUI — chat, plan, edit workflows, and automate tasks with local or cloud LLMs.

A lightweight ComfyUI node that intelligently rounds image dimensions to compatible resolutions using smart reflection mirroring.

A node pack for ComfyUI

Ollama in ComfyUI

NODES: ORvTextEncoderGoogleEmbeddingGemma3, ORvEmbeddingsHeatmap, ORvEmbeddingsSpectrogram, ORvStringConsoleDebug

A simple counter, when pressing 'Queue Prompt' resets the count.

General-purpose ComfyUI nodes: an LLM prompt enhancer for the Allma backend (multi-image, audio, LoRA metadata, presets), plus graph utilities — a branch muter, a named data bus, a VRAM unloader, a combo mirror and a live-text display.

A collection of custom node for online chat with APIKey to OpenAI ChatGpt4xx... Grok ....

Wildcard nodes compatible with any form of CLIP-text encode, with recursive HTML tags, weights, and adjoining support.

NovelAI V5 generation and editable PNG/WebP metadata restoration nodes for ComfyUI.

Utility nodes for prompt engineering with random selectors, dynamic note-taking, advanced image saving with metadata, and text debugging tools.

This extension adds helper nodes for looping over prompts and other data types.

Local MCP bridge for inspecting and editing a live ComfyUI canvas.

ComfyUI node to upload images to the Trinetra image host, with rate-limit-aware backoff, an expiry picker, and a live folder dropdown.

utilitaires pour ComfyUI, conçue pour effectuer des calculs et des conversions

This ComfyUI node allows you to extract all images found in various types of documents and save them to disk. It also provides a preview of the first extracted image.

This repository contains a versatile custom node for ComfyUI, Display Image with Mask, designed to offer advanced image viewing, masking, and saving capabilities directly within your workflow.

ComfyUI integration for Mistral LLM language model. (Description by CC)

VFX-oriented nodes for ComfyUI: HDR LogC3 / LogC4 decode, EXR I/O, and an interactive ACEScct color grading panel powered by exr-viewer.

Helps with tha chess peices

ComfyUI plug-in collection of basic translator, LLM translator and a series of LLM Service Connectors.

ComfyUI plug-in collection of basic translator, LLM translator and a series of LLM Service Connectors.

Nodes:Caching Image Loader.

A minimal ComfyUI custom node package for converting images to 3D meshes using TripoSR and rendering them from 8 cardinal/intercardinal directions.

Image-only switch for ComfyUI that checks connected inputs from top to bottom and outputs the first available image. (Description by CC)

Pack and unpack any number of typed values through a dynamic pipe in ComfyUI.

ComfyUI custom node for merging multiple text or tag lists with deduplication and custom delimiter support. (Description by CC)

Rotate RGB or RGBA images and masks at arbitrary angles with configurable scaling and backgrounds in ComfyUI.

Disable API Templates (partner templates or paid templates) in ComfyUI