ComfyUI Custom Extensions

ComfyUI-PosterCraft is now available in ComfyUI, PosterCraft is a unified framework for high-quality aesthetic poster generation that excels in precise text rendering, seamless integration of abstract art, striking layouts, and stylistic harmony.

A simple 3D model processing tool within ComfyUI.

Ready to use upon download. No need to install dependencies for the time being. If there are new functions or suggestions, please provide feedback. Attention! The delfile node is not recommended for use on servers. I am not responsible for any losses incurred.

ComfyUI-MagnifyGlass: A powerful & customizable magnifying glass for ComfyUI. Zoom into canvas details with smooth controls, configurable activation, custom styles (shape, size, border) & WebGL performance.

tool set for developing workflow and publish to web api server

a node for [a/AuraSR](https://github.com/fal-ai/aura-sr)

ZImage Turbo LoRA Stack V4 custom node for ComfyUI.

A ComfyUI extension that allows you to find and locate nodes by their ID, title, pack, or type in complex workflows.

Seamless tiled refinement and upscaling. Refine an image tile by tile with no visible seams, or refine only a masked region of a large image without processing the rest. Each tile is conditioned on its already-refined neighbors held as frozen context. The VL nodes (Krea 2) need no prompt at all: each tile's conditioning is a slice of one whole-image vision encode, so prompt objects never reappear in the wrong tile. Works with any guider, including NAG (Normalized Attention Guidance).

wrapper for the TLB-VFI: Temporal-Aware Latent Brownian Bridge Diffusion for Video Frame Interpolation project

A ComfyUI-native FLUX.2 ControlNet toolkit for FLUX.2-dev Fun ControlNet Union checkpoints, with unified dense BF16 and mixed FP8/BF16 loading, single-ControlNet Apply nodes, flat non-recursive multi-ControlNet orchestration, reference-image conditioning, latent caching, and experimental mask-aware in/out-paint support while preserving ComfyUI's native FLUX.2 sampler and model-management lifecycle.

DiffusionLight (Turbo) implemented in ComfyUI

This node creates a sequence of frames by moving and scaling a subject image over a background image.

Standalone torch.compile optimization node for ComfyUI with speed-focused presets

ComfyUI custom nodes collection with workflow utilities, image processing, and AI integration

Utility nodes for interacting with images and image sizes

Prompt manager, prompt crafter, visual prompt

A comprehensive ComfyUI node collection that integrates with the HiTem3D API to generate high-quality 3D models from images, featuring advanced HTML preview capabilities and dynamic content generation.

A plugin application that utilizes ComfyUI to generate 360-degree panoramic images. It primarily works by converting between flat images and equidistant cylindrical projections, stretching flat images into a curved format, and setting regional conditions, while harnessing the power of large models to produce 360-degree panoramic views.

Image Browsing: browsing, download and delete.

Data analysis custom modules for ComfyUI - Use Pandas & Matplotlib from within ComfyUI

Create Video datasets straight from YT or a video file path

ComfyUI Utility Nodes by Imagineer. 1: Catch and Edit Text; useful for grabbing AI generated prompts which you edit by hand. Doing so mutes the upstream node, improving speed and saving external calls and budget. <br>2. Preview Image - No Save: Previews as they should be

A practical plug‑in that adds a StableStudio style user interface to ComfyUI. This project aims to give you a clean, responsive UI for managing StableStudio features inside ComfyUI, without complicating your workflow. It focuses on reliability, speed, and a calm, uncluttered design that helps you work faster.

Basic audio tools using torchaudio for ComfyUI. It is assumed to assist in the speech dataset creation for ASR, TTS, etc.

A manga page creation SPA running on top of ComfyUI. Manage pages by 'works' (page groups), create pages from templates, and place images, speech balloons, text, shapes, and 3D poses into panels. Includes a layer-based image editor (layers, masks, adjustment layers), font manager with reusable style/preset system, AI image generation (Nanobanana/Gemini), script/storyboard management, and export to JPEG/PNG/WebP/PDF/EPUB. Multi-language UI (ja/en/zh).

Nodes:Apply CDTuner, Apply Negapip. This extension provides the [a/CD(Color/Detail) Tuner](https://github.com/hako-mikan/sd-webui-cd-tuner) and the [a/Negative Prompt in the Prompt](https://github.com/hako-mikan/sd-webui-negpip) features.

MiniMax H3 Chain Director (CN/EN): auto split long video into r2v+i2v segments with first-frame locking, stitch video+audio in one node. 链式导演台：自动分段拼接 MiniMax H3 长视频

A dependency-free ComfyUI custom node for controlling the execution order of nodes by routing any type/number of data through it, ensuring all input connections resolve first before being passed to outputs.

Timeline-based director node for MiniMax H3 — visual CUT/segment editor, multi-character reference images, reference video + reference audio (with per-clip audio pairing for lipsync/voice reference), First/Last Frame mode, and automatic chunking with a hybrid hard-lock continuation option for videos longer than a single H3 generation.

Triton kernel acceleration + W8A8 INT8 quantization for Z-Image S3-DiT in ComfyUI

The Olm LUT is a custom node for ComfyUI that allows you to apply .cube LUT (Look-Up Table) files to images within your generative workflows. It supports creative workflows including film emulation, color grading, and aesthetic stylization using LUTs.

ComfyUI Assistant is your one stop plugin for everything you need to get started with comfy-ui. Now it provides useful courses, tutorials, and basic templates.

SenseNova-U1 custom nodes for ComfyUI (API + local inference).

Custom ComfyUI nodes for QwenImage ControlNet achieving 100% output compatibility with VideoX-Fun's diffusers pipeline while leveraging ComfyUI's efficient model loading system.

A ComfyUI model loader that uses the fastsafetensors library to perform very fast, zero-copy loading from storage to VRAM.

A high-performance image upscaling plugin for ComfyUI with CUDA acceleration, multi-GPU support, mixed precision, and Tensor Core optimization

节点工具箱、视频图像批处理、JSON数据处理

A long-form video generation suite that chains keyframe images into coherent videos with automatic segment transitions and cross-dissolves. (Description by CC)

ComfyUI node: extract fps/frames/duration/width/height from video

Node to tryoff clothes

This custom node helps to transform latent in different ways.

[DEPRECATED — use ComfyUI_Eclipse instead] This node contains a lot of small little helpers like switches, passers and selectors that i use a lot to build my workflows.

ComfyUI nodes for real-time use cases

ComfyUI custom nodes for FireRedTTS3: zero-shot voice cloning (24 languages + 21 Chinese dialects), voice design, and speech editing, with native inference and AIMDO DynamicVRAM support

Generate sets of tiles from images or from prompts

This node allows downloading models and other resources used within a ComfyUI workflow making it easier to resolve referenced assets and share workflows

A powerful ComfyUI custom node that brings Google's Gemini TTS capabilities directly to your workflow. Generate high-quality speech with 30+ voices supporting both free and paid tiers.

Custom nodes for interpolating between, well, everything in the Stable Diffusion ComfyUI.

ComfyUI implementation of Regional Adaptive Sampling, (original implementation at https://github.com/microsoft/RAS).

A comprehensive VRAM management tool for ComfyUI. Includes automatic cleanup and GPU monitoring.

ComfyUI V3 nodes for FireRedAudio using an isolated Transformers 5.8 worker

A quick and easy ComfyUI custom node for setting SDXL-friendly aspect ratios.

Learned MiniMax H3 latent upscaling with integrated low-sigma refinement for ComfyUI.

Inference-time techniques to enhance diffusion-based image generation quality through random search and zero-order optimization algorithms

ComfyUI-HiggsAudio is now available in ComfyUI, Higgs Audio v2 is a text-audio foundation model from Boson AI.

Building upon the excellent foundation of ComfyUI_TensorRT, this version focuses on full support for the Anima architecture and advanced TensorRT acceleration. It aims to push NVIDIA RTX™ GPU performance to its limits for the next generation of Stable Diffusion workflows.

A collection of nodes for common tools, including text preview, text translation (multi-platform, multi-language), image loader, webcamera capture.

🌐 Live Search: Real-time web search and AI summarization. Internet search, web scraping, fact checking, weather, news, GPS coordinate conversion. DuckDuckGo search with GPT-5.1, DeepSeek-V3, Gemini 3 Pro, Claude 4.5, Qwen3, Doubao, SiliconFlow (69+ models), Llama 4, Ollama. Modular architecture with API loader. Toggle web search on/off for pure LLM mode.

Nodes: StreamRecorder, VideoFormatConverter, ZZX_PaintsUndo

Use model bending to push your model beyond its visuals limits. These nodes allow you to interactivly manipulate the low-level activations of models (e.g., UNet, VAE, LoRAs) to create creative variations that are hard to achieve with other methods

Strict official-style modulation guidance node for Anima models in ComfyUI.

ComfyUI custom node for image tagging using WD v3 models, based on .

Portrait Tools: Facial detection cropping, alignment, ID photo, etc.

A custom node for Stable Diffusion ComfyUI to enable easy selection of image resolutions for SDXL SD15 SD21

Audio processing nodes for comfyui.

ComfyUI dust3r

One-node MiniMax H3 prompt engineering, semantic asset references, validation, and official conditioning.

Send images generated by ComfyUI to Eagle image management software

ComfyUI memory cleanup nodes for VRAM and RAM management.

LM Studio ComfyUI integration with easy model selection and many optimizations! Features auto model discovery, vision/VLM support, reasoning extraction for DeepSeek R1/Qwen3/QwQ models, and detailed stats with VRAM management.

ComfyUI nodes for Qwen-Image-Edit and Qwen-Image — 30 nodes for editing, generation, multi-stage, LoRA, Spectrum acceleration, ControlNet-guided generation, prompt rewriting, and 2× upscale VAE

An unofficial Python library for [a/Suno AI](https://www.suno.ai/) API

SaltAI AudioViz contains ComfyUI nodes for generating complex audio reactive visualizations

This is an image recognition node for ComfyUI based on the RAM++ model from [a/xinyu1205](https://huggingface.co/xinyu1205). This node outputs a string of tags with all the recognized objects and elements in the image in English or Chinese language. For image tagging and captioning.

These are ComfyUI nodes to assist in converting an image to sketches or lineArts.

Unofficial ComfyUI implementation of [a/UniWorld-V1](https://github.com/PKU-YuanGroup/UniWorld-V1).

Zero-1-to-3: Zero-shot One Image to 3D Object, unofficial porting of original [Zero123](https://github.com/cvlab-columbia/zero123)

These are companion nodes for Texturaizer, a Blender plugin that connects complex 3D data to ComfyUI. https://texturaizer.com

Advanced nodes of the Level Pixel company (levelpixel, LP). Includes convenient advanced nodes for working with LLM и VLM models (LLaVa) with GGUF format. Qwen2.5-VL and Qwen2.5 supported. Also included is a node for the RAM model. Nodes have the ability to automatically unload models from VRAM.

Automatic background removal and transparency generation for ComfyUI

DWPose & NLF skeleton editor, retargeter, and renderer for ComfyUI. Extract body/hand/face keypoints (+ optional NLF 3D) from video, edit them frame-by-frame in a full-screen timeline + dope sheet + graph + 3D orbit editor with animated camera, retarget DWPose proportions per-cluster, and render clean pose images. Outputs standard ComfyUI types — drives WanAnimate, ControlNet, SCAIL, LTX-Video, UniAnimate, and any other pose-driven pipeline.

Solricks is an Istanbul-based technology brand focused on AI systems, custom software, cloud infrastructure, hardware, and enterprise technology solutions.

Neons Style Explorer — a hand-written style-prefix catalog for ComfyUI, in natural language or booru tags, with a preview gallery.

A smart, customizable resolution selector node with live aspect ratio preview, image overlays, and checkerboard support — cleanly parsed from a human-readable text file.

Nodes: AstropulsePixelDetector, BackgroundRemover, ImagePadForBetterOutpaint, InpaintingPipelineLoader, Inpainting, ...

A ComfyUI node for enhancing prompts using various LLM providers

A dual-model image-to-video sampler for ComfyUI that combines high-noise and low-noise models in a single node, replacing two KSamplerAdvanced nodes with zero overhead and bit-identical frame generation. (Description by CC)

Comfy-MSS provides ComfyUI nodes for music source separation powered by [pymss](https://github.com/pymss-project/pymss). It supports both built-in pymss model catalog entries and user-provided MSST models, allowing workflows to separate an input audio track into stems such as vocals, instrumental, drums, bass, and other sources.

DLSS 5 Neural Rendering (NGX feature 18) enhance/upscale plus DLSS frame generation (2x/3x/4x frame-rate multiplication) for ComfyUI IMAGE and VIDEO. Windows in-process D3D12 bridge and Linux/Wine worker backends.

Latent space walks for latents, conditionals and noise

This node is based on MykolaL/StableDesign

This node is based on MykolaL/StableDesign

A ComfyUI custom node package that allows downloading and organizing Instagram content directly in your ComfyUI Output folder

Nodes: CLIPTextEncodeA1111, RerouteTextForCLIPTextEncodeA1111.

A ComfyUI wrapper implementation of ThinkSound - an advanced AI model for generating high-quality audio from text descriptions and video content using Chain-of-Thought (CoT) reasoning.

Transparent WebM overlay for live previews and sampler previews in ComfyUI.

Flux Pro via Replicate API Create API key at [a/https://replicate.com/account/api-tokens](https://replicate.com/account/api-tokens) Copy config.ini.example to config.ini and put the replicate key there.