ComfyUI Custom Extensions
Browse 9,420 ComfyUI extensions and 73,215 custom nodes.
Bridge between ComfyUI and blender's ComfyUI-BlenderAI-node addon.
Symbolic Music Generation, NotaGen node for ComfyUI.
ComfyUI extension for FLUX FaceIR aligned-face and whole-image restoration.
A custom node for ComfyUI that provides a visual gallery for managing and applying multiple LoRA models.
Integrates Polotno Canvas Editor into ComfyUI for advanced image editing and design
ComfyUI custom node implementation of VideoMaMa for video matting with mask conditioning.
ComfyUI node for Maya1 TTS - Expressive voice generation with 20+ emotions, voice design, and SNAC neural codec
Nodes:Stability SD3, Stability Outpainting, Stability Search and Replace, Stability Image Core, Stability Inpainting, Stability Remove Background, Stability Creative Upscale. Add API key to environment variable 'SAI_API_KEY' Alternatively you can write your API key to file 'sai_platform_key.txt' You can also use and/or override the above by entering your API key in the 'api_key_override' field of each node.
ComfyUI V3 nodes for the official IndexTTS 2.5 model, by T8star-Aix
Generate Audio from any video and or text
Optional NSFW content expansion for ComfyUI-Photoreal-Prompt-Builder with explicit pose, action, and group action expansions.
ComfyUI custom nodes for the Civitai Orchestration API — run Civitai recipes (image/video/audio generation, training, analysis) from inside ComfyUI
Two simple nodes for stereoscopic image generation. Nodes: Stereo Image Node - a basic port from the Automatic1111 stereo script in thygate/stable-diffusion-webui-depthmap-script, LazyStereo - a naïve stereo image generator
Custom set of nodes used by glif.app. With glif you can build mini apps that are powered by custom comfy workflows.
ComfyUI nodes for MiniMax H3 loading, directing, prompting, acceleration, RTX and video workflows
An all-in-one node for ComfyUI that implements AceStep 1.5 SFT (Supervised Fine-Tuning), a high-quality music generation model. This node replicates the full functionality of the official Gradio pipeline, offering fine control over audio synthesis parameters.
ComfyUI OmniConsistency Nodes is a collection of nodes for ComfyUI that allows you to load and use OmniConsistency models.
A custom node for ComfyUI that integrates LM Studio's vision models to generate text descriptions of images. It provides a flexible and customizable way to add image-to-text capabilities to your ComfyUI workflows, working with LM Studio's local API.
ComfyUI-LightVAE is a collection of LightX2V VAE custom nodes designed for ComfyUI, supporting high-performance video VAE models including LightVAE and LightTAE.
This extension provides a node that allows you to overlay text on an image or a batch of images with support for custom fonts and styles.
A [a/Matchering](https://github.com/sergree/matchering)-node for ComfyUI. NOTE: You take TWO audio files and feed them into Matchering
ComfyUI implementation of the [a/InstantX IP-Adapter for SD3.5 Large](https://huggingface.co/InstantX/SD3.5-Large-IP-Adapter).
ComfyUI_LucidFlux:Caption-Free Universal Image Restoration with a Large-Scale Diffusion Transformer,you can use it in ComfyUI
This is a custom node for ComfyUI that allows you to use the WaveSpeed AI API directly in ComfyUI. WaveSpeed AI is a high-performance AI image and video generation service platform offering industry-leading generation speeds. For more information, see [a/WaveSpeed AI Documentation](https://wavespeed.ai/docs).
Batched Runge-Kutta Samplers for ComfyUI
A single-node ComfyUI custom package that turns up to 4 reference images into a 2×2 stitched grid and injects it into Flux Klein 9B conditioning.
ComfyUI 图像生成链路中的可视化编辑与桥接系统,支持遮罩绘制、涂鸦、裁切、扩图与调色。
Tile-based refinement nodes for ComfyUI, built around the Hildegard-Refiner reference-latent scheme. Designed for FLUX.2 Klein.
A facefusion custom node for ComfyUI. Swap or restore faces for image or video
ComfyUI custom node for Attention Couple
ComfyUI-CrewAI aims to integrate Crew AI's multi-agent collaboration framework into the ComfyUI environment. By combining the strengths of Crew AI's role-based, collaborative AI agent system with ComfyUI's intuitive interface, we will create a robust platform for managing and executing complex AI tasks seamlessly
Image to caption with CLIP ViT-L/14. Small and fast addition to the CLIP-L model you already have loaded to generate captions for images within your workflow.
An extension supporting more streamlined prompt queue management.
HyperFlow: 8-step LoRA + two-time (t, r) conditioning for MiniMax-H3
Professional voice cloning and audio editing node for ComfyUI using Step Audio EditX
Progressive upscaling node for flow-matching models that incrementally doubles resolution with flow-consistent re-noising, denoising, and skip residual blending to preserve composition.
Comfyui-deepseek-r1 Node Plugin
a custom node for [a/MaskGCT](https://github.com/open-mmlab/Amphion/blob/main/models/tts/maskgct/README.md) to Zero-Shot Text-to-Speech
This custom node for ComfyUI allows you to create layers of an image based on input masks and save them into a PSD file.
Provides nodes to utilise NVIDIA NIM, a set of accelerated inference microservices that allow you to run AI models on NVIDIA GPUs anywhere.
This node uses YOLO model to obtain bboxes required by ComfyUI's official SDPose node for multi-person processing. (Description by CC)
Kandinsky S2V 1.0 Lite and Pro distilled speech-to-video (talking avatar) models for ComfyUI
Process text and images with GGUF LLMs in ComfyUI using llama.cpp, including Qwen3-VL, Qwen3.5, Qwen3.6, Gemma 4, and gpt-oss.
Run Qwen3-VL, Qwen3.5, and Qwen3.6 GGUF models in ComfyUI with llama.cpp.
LoRA training and inference apps for ComfyUI: open /zentrainer/ after installing.
A set of custom nodes for ComfyUI that leverage the Gemini 2.5 Flash Image Preview API to generate images from text prompts, single images, and multiple images.
Modularized version of Disco Diffusion for use with ComfyUI.
ComfyS3 seamlessly integrates with [a/Amazon S3](https://aws.amazon.com/en/s3/) in ComfyUI. This open-source project provides custom nodes for effortless loading and saving of images, videos, and checkpoint models directly from S3 buckets within the ComfyUI graph interface.
ComfyUI Wrapper for UltraTex: Unleashing 2K Multi-View Diffusion for 3D Texturing
Qwen Image Edit uses a Vision Language Model to analyze input images and automatically enhance prompts with detailed descriptions.
Qwen Image Edit uses a Vision Language Model to analyze input images and automatically enhance prompts with detailed descriptions.
FL Song Gen - AI-powered song generation nodes for ComfyUI. Generate complete songs with vocals and instrumentals from lyrics using Tencent's SongGeneration (LeVo) model. Features style transfer, auto style presets, dual-track output, and up to 4m30s song generation.
Visualize your textures inside ComfyUI
This is a Seed-X-PPO-7B ComfyUI plugin. Easy to use
Musubi Tuner by kohya_ss
Cosmos-style rectified-flow sampler nodes for Anima models in ComfyUI.
High‑quality music audio enhancement for ComfyUI: FlashSR super‑resolution + Fat Llama spectral enhancement (GPU & CPU).
KV-Edit: Training-Free Image Editing for Precise Background Preservation,you can use it in comfyUI
The custom Snap processing node has been converted for ComfyUI production. It currently includes a simple PyQt5 interactive interface as well as practical nodes for basic operations like area calculation..
This project is an unofficial ComfyUI implementation of [a/Video Depth Anything](https://github.com/DepthAnything/Video-Depth-Anything), for depth estimation on long videos without compromising quality, consistency, or generalization ability.
a custom nodde for [a/IMAGDressing](https://github.com/muzishen/IMAGDressing)
Steerable, resumable long-form MiniMax H3 video for ComfyUI
The ComfyUI-fal-Connector is a tool designed to provide an integration between ComfyUI and fal. This extension allows users to execute their ComfyUI workflows directly on [a/fal.ai](https://fal.ai/). This enables users to leverage the computational power and resources provided by fal.ai for running their ComfyUI workflows.
A Chinese and English text translation plugin for ComfyUI.
This is a PuLID node that has been extended with new features.
This project provides a node for ComfyUI to use the JoyCaption-Beta model in GGUF format for image captioning.
Original repo: [a/https://github.com/luckyhzt/LVCD](https://github.com/luckyhzt/LVCD)
Nodes:TileSplit, TileMerge.
'🔢 Prompt Combinator' is a node that generates all possible combinations of prompts from several lists of strings. '🔢 Prompt Combinator Merger' is a node that enables merging the output of two different '🔢 Prompt Combinator' nodes.
Side by Side Stereoscope node for ComfyUI by SamSeen
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, automatic chunking, two-stage sampling, a VAE re-encode continuity mode that hard-freezes a real re-encoded window of the previous chunk's output across the chunk boundary, and an explicit Seed Hunt on/off switch with a 1-4 candidate count. Bundled with its companion Muse Minimax Refine node — picks one Seed Hunt candidate and continues its sigma schedule at a higher resolution, including full multi-chunk Latent-Only Scouting candidates, not just a single chunk.
I wanted a node to save my prompts and optionally take in an external prompt from a llm and save it, didn't see one, so I made it.
Seamless looping video with MiniMax H3 in ComfyUI: circular padding, loop decode, ControlNet restyling and loop-safe interpolation.
ComfyUI_MangaNinjia is a ComfyUI node of MangaNinja which is a Line Art Colorization with Precise Reference Following method.
A comprehensive collection of ComfyUI custom nodes that replicate the functionality of popular Nuke compositing nodes. Includes merge, grade, transform, blur nodes and more for professional compositing workflows.
IP-Adapter custom node for Anima in ComfyUI that injects reference image features via decoupled cross-attention for character-consistent generation.
ComfyUI toolkit with custom nodes for BBox canvas visualization, image brush mask drawing, and coordinate annotation for debugging and external data bridging. (Description by CC)
Prompt Generator for Video, Audio, Image, and Text. A node for ComfyUI. Including Deepseek, Alibaba Cloud Qwen, Google Gemini, and locally selected models, etc.
Custom node packs for ComfyUI: FolderIncrementer (auto-versioning) + MaskEditControl (pinpoint mask editing with SAM 2.1 / SAM 3, per-axis erode/expand, point editor, bbox tools, video mask propagation).
让 ComfyUI 像 WebUI 一样写提示词:中文翻译、Tag 自动补全、LoRA 管理、区域提示词控制,支持 Anima 与 SDXL 工作流。
SIFT-based image alignment, intelligent mask cropping and restoration for ComfyUI
implementation of [a/AniPortrait](https://github.com/Zejun-Yang/AniPortrait) generating of videos, includes self driven, face reenacment and audio driven with a reference image
Dynamic Band Splitting (DBS) for ComfyUI: separate a soundtrack into music, sound effects and speech with the official DnR v2 models (DBS-TIGER 1.0M / DBS-BSRNN 4.7M parameters; CPU or CUDA). Same output as the official package, bit for bit on the CPU. Remove or keep the dialogue of a video.
ComfyUI extension for video-based prompting and processing with support for various models and video processing capabilities
This node allows using Hugginface remote server for latent decoding. Currently supported models: SD, SDXL, Flux, HunyuanVideo
The node has been created through vibe coding. The main objective is for identity consistency for FLUX.2 klein 9b models in ComfyUI.
This project is the ComfyUI implementation of the TaylorSeer project [a/https://github.com/Shenyi-Z/TaylorSeer](https://github.com/Shenyi-Z/TaylorSeer).
Adds a global /workflow/convert API endpoint to convert non-API workflow formats to API format for ComfyUI execution
A collection of AI-powered photography and image generation tools for ComfyUI, featuring the SID_AIPromptGenerator node that uses Anthropic's Claude API to analyze images and generate optimized prompts for various image generation models.
Experimental native Qwen-Image-2.1 block cache, spectral forecasting, and attention adapters.
Wrapper ComfyUI integration for the [a/Flux Omini Kontext](https://github.com/Saquib764/omini-kontext) pipeline, enabling seamless character/object insertion into scenes using FLUX.1-Kontext-dev with LoRA adaptation.
This custom node is designed to provide graphical documentation for ComfyUI custom nodes.
DeepExtractV2 – lightning-fast, high-quality audio separator. Instantly isolate vocals, drums, bass, and more from any track. Ideal for creators, DJs, and audio pros.
Nodes: LoadImageFromPath. Load Image From Path loads the image from the source path and does not have such problems.
Removes the ComfyUI proxy middleman entirely, allowing direct API calls to vendors with your own keys, rates, and data privacy.
Decor8 AI Virtual Staging, Interior Design, and Design Inspirations (Requires API key from www.decor8.ai). Requires a valid pay-as-you go subscription plan or credits bundle. Reach out to [email protected] if you have any questions or need test credits.
Moondream's gaze detection feature wrapper node.
InstantHMR real-time human mesh recovery for ComfyUI: MHR body pose, shape, keypoints and mesh for every person, on the native SAM 3D Body nodes.
A minimalist node that calls LLMs, combined with one API, can call all language models, including local models.