ComfyUI Custom Extensions

ComfyUI nodes for torch.compile.

ComfyUI-MoviiGen is now available in ComfyUI, MoviiGen 1.1 is a cutting-edge video generation model that excels in cinematic aesthetics and visual quality.

Make Step1X-Edit avialbe in ComfyUI.

Control and visualization toolkit (Gatekeeper, Slider) and more in the future.

Custom node for sending emails via SMTP

Professional 3D face reconstruction for ComfyUI using the Pixel3DMM method

a comfyui node for viewing Live2D model

Faithful volumetric video of real performances — surround coverage from multi-camera capture, engineered for measurable stability.

Implementation of VIBE (Visual Instruction Based Editor) as a custom node for ComfyUI that enables image editing using natural language instructions leveraging Sana1.5 and Qwen3-VL models.

Custom samplers and schedulers for ComfyUI

A ComfyUI custom node for loading images from a contact sheet of recent files

A ComfyUI custom node that provides a database-driven prompt management system. Store, organize, and edit prompts in categories with persistent JSON storage.

Create a custom scheduler from a weighted average of the built-in schedulers

image and latent quilting nodes for comfyui

Adds extra API functionallity and prometheus endpoint

ComfyUI nodes that load a Krea2 LoRA, strip selected layer groups from memory and apply the remaining weights to the diffusion model, with simple and advanced versions. (Description by CC)

Dependency-free ComfyUI skin-tone correction nodes with bilingual UI and optional MediaPipe masks

This node pack helps to convert various files to Markdown. It supports pdf, pptx, xlsx, docx, html and image files.

Optimize prompts for text-to-image models at no cost.

Pixar's OpenUSD pipeline integration for ComfyUI. Natively compose and render 3D scenes within your node workflows.

Simple stack push/pop style nodes for images, strings, integers and generic objects (image batches, latents, face models etc)

ComfyUI_Dados_Nodes is a collection of custom nodes for ComfyUI, designed to enhance functionality and provide integration with various services, including Pinterest. This privacy policy explains how these nodes handle user data. NOTE: [a/privacy_policy](https://github.com/dadoirie/ComfyUI_Dados_Nodes/blob/master/privacy_policy.md)

High-fidelity, architecture-aware LoRA fusion for Z-Image (S3-DiT). Features simultaneous TIES-merging, per-block tuning, and no-loss export.

A collection of custom nodes for ComfyUI that integrates various Vision-Language (VL) models, including Xiaomi MiMo-VL, LiquidAI LFM2-VL, Kwai Keye-VL, AIDC-AI Ovis2.5 and Ovis-U1. Tested with models: AIDC-AI/Ovis2.5-2B, AIDC-AI/Ovis2.5-9B, Kwai-Keye/Keye-VL-8B-Preview, MiMo-VL-7B-RL-GGUF, LiquidAI/LFM2-VL-450M, LiquidAI/LFM2-VL-1.6B, AIDC-AI/Ovis-U1-3B.

Suite of native custom nodes for high-precision multi-frame storyboard generation, structural isolation, frame extraction, and cinematic tuning. (Description by CC)

A powerful workflow scheduling extension for ComfyUI that enables automated daily execution of workflows with an intuitive web interface ,Adding shutdown computer after workflow node

Modular procedural prompt generation — wildcards, fixed values, combines, derivations and constraints resolved in a context pipeline, with a manager UI for a reusable module library.

A custom ComfyUI node that implements A1111-style prompt handling with proper isolation and emphasis math.

Advanced PDF processing, OCR, Image and text parsing, smart image crop nodes for ComfyUI. Features include multi-language OCR (Surya, PaddleOCR VL), AI vision analysis (Florence-2, LayoutLMv3), advanced PDF extraction with quality assessment, spread detection for scanned books, and comprehensive layout analysis. Supports 90+ languages with multiple output formats.

Nodes that implement functionality similar to the Dynamic Prompts extension for A1111.

Text-to-3D and Image-to-3D generation using Tencent Cloud Hunyuan 3D Global API. Supports PBR materials, face count control (40K-1.5M faces), and multiple generation types (Normal/LowPoly/Geometry/Sketch). Outputs industry-standard GLB format. Requires Tencent Cloud account with API access.

NODES:Nsfw Image Check Node

Original repo: [a/https://github.com/lllyasviel/IC-Light](https://github.com/lllyasviel/IC-Light) Models: [a/https://huggingface.co/lllyasviel/ic-light/tree/main](https://huggingface.co/lllyasviel/ic-light/tree/main), [a/https://huggingface.co/digiplay/Photon_v1/tree/main](https://huggingface.co/digiplay/Photon_v1/tree/main) models go into ComfyUI/models/unet

FL FishSpeech - AI Text-to-Speech & Voice Cloning for ComfyUI. High-quality 44.1kHz speech synthesis with zero-shot voice cloning using OpenAudio S1-mini (Fish Audio). Features DualAR Transformer, DAC codec, emotion control tags, and built-in Whisper transcription.

Unofficial ComfyUI implementation of Self-Guidance.

A custom node for ComfyUI to create a prompt based on a list of keywords saved in CSV files.

Unofficial optimized and enhanced fork of [a/LatentSync 1.5](https://github.com/bytedance/LatentSync) implementation for ComfyUI on Windows and WSL 2.0.

Nodes:GMIC Image Processing.

Tiled upscale and refine for reference-latent edit models (FLUX.2 klein and similar): regenerate detail beyond the resolution a model can handle in one pass, without visible seams.

y.k.'s comfyui node suite

WorkflowX bundles ComfyUI configuration profile nodes with XFlows workflow management, XPrompts prompt libraries, XNodes reusable node snippets, and the JsonX prompt system.

ComfyUI zip package image processing

A faster tiled VAE decode node for ComfyUI based on LTXV, optimized for low VRAM scenarios with configurable tiling, progress bars, and improved logging for decoding large images and videos.

The PonySwitch node is a custom node for ComfyUI that modifies prompts based on a toggle switch and adds configurable pony tags.

This initiative represents a solo venture dedicated to integrating a stereopsis effect within ComfyUI (Stable Diffusion). Presently, the project is focused on the refinement of node categorization within a unified framework, as it is in the early stages of development. However, it has achieved functionality in a fundamental capacity. By processing a video through the Side-by-Side (SBS) node and applying Frame Delay to one of the inputs, it facilitates the creation of a stereopsis effect. This effect is compatible with any Virtual Reality headset that supports SBS video playback, offering a practical application in immersive media experiences.

Save and load MiniMax H3 AV NestedTensor latents, avoiding the built-in Save Latent limitation for H3 video/audio latent pairs.

High-performance custom node for LTX2 video generation designed to reduce GPU usage, VRAM requirements, and temperature on consumer GPUs.

Publish a ComfyUI generation to Flow Production Tracking as a Version, with the model, prompt, seed and workflow that made it, and load a Version's media back into a graph.

A ComfyUI node set for creating and managing visual storyboards. Includes nodes for markdown rendering, image composition, and storyboard sequence management.

Simple checkpoint loader for TaiyiXL

ComfyUI Mini Toolkit – Helps reduce node redundancy. Let’s see if there’s time to keep it updated regularly.

A combined node for ComfyUI with six configurable tabs for managing multiple parameter configurations, including guidance, sampler, scheduler, steps, and denoise.

Generate Stable Diffusion prompts using Qwen3-8B via Ollama with 7 style presets (cinematic, anime, photorealistic, fantasy, abstract, cyberpunk, sci-fi)

A professional ComfyUI custom node for accurate text-audio alignment using [a/WhisperX](https://github.com/m-bain/whisperX).

Professional photo editing tools for ComfyUI. Photoshop-style curves, levels, and color adjustments, AI-powered masking / background removal, LUT color grading, advanced channel masking, color management, frequency separation, blur and noise/grain filters, inpaint mask tools, and workflow utility nodes.

ComfyUI custom node for tagging anime images using TIMM models from animetimm, supporting general tags, character tags, artist tags, and rating tags with batch image processing.

Copy the shot, not the actors: reference video → depth map, OpenPose skeleton (DWPose, tracked and smoothed) or canny line video for Seedance 2.0/2.5 reference video, MiniMax H3 Fun ControlNet (depth / pose / canny) and Wan VACE. Six nodes: Depth / Pose / Canny as IMAGE→IMAGE and VIDEO→VIDEO with the model presets (fps, frame-size multiples) applied. Apache-2.0.

Automatically enables TensorFloat-32 (TF32) acceleration for NVIDIA RTX 30/40/50 series GPUs in ComfyUI.

A training-free ComfyUI model patcher that accelerates diffusion sampling via Chebyshev spectral forecasting, compatible with standard samplers like Euler and DPM++. (Description by CC)

Enhance ComfyUI workflow by providing powerful node search, attribute search, and port navigation features with Ctrl+F shortcut support.

Custom ComfyUI nodes using CLIP Vision models for image similarity analysis.

Custom nodes for Motif-Video 2B (Motif Technologies) text-to-video and image-to-video diffusion model. Includes model loaders, samplers, and a workflow-driven automatic model download dialog.

A single ComfyUI node for executing arbitrary Python code with arbitrary inputs of any types.

collection of SVG manipulation tools

Custom node to manage prompts in YAML format.

YAMLでプロンプトを管理するノード/Custom node to manage prompts in YAML format.

Near implementation of Differential Correction in Wavelet domain (DCW) for SNR-t bias correction. Based on arXiv:2604.16044 (Yu et al., 2026). No external dependencies. Compatible with SDXL, SD1.5, Flux, Cosmos, EDM, DiT. (README in Korean)

This extension applies a patch that limits the model loading speed when using an HDD in a Windows environment. See [a/comfyanonymous/ComfyUI#1992](https://github.com/comfyanonymous/ComfyUI/issues/1992). [w/As this patch is only useful in very limited environments, its installation is not recommended under normal circumstances. Memory usage may increase.]

www.nuvulabs.ai | Install missing models & custom nodes, production ready VFX & film subgraphs and workflows for secure offline generation, and dependency installations including SageAttention and Triton.

A collection of high-quality nodes for ComfyUI, dedicated to improving the sharpness, clarity, and texture of your images

Orion4D Anaglyph is a high-performance custom node designed to transform 2D images into stereoscopic renders via a depth map. It offers total control over parallax, convergence, and depth processing to ensure optimal visual comfort

HSL-based color consistency node for ComfyUI. Supports independent matching of Luminance, Hue, and Saturation with mask protection and bilingual UI.

A lightweight ComfyUI custom node that applies a direct (1,12) delta patch to diffusion_model.txtfusion.projector.weight, with editable presets, custom value input, and a LoRA-style strength control.

Drop-in workaround for LiteGraph-mode widget width bug in ComfyUI frontend. (Description by CC)

Add a button to the menu to save and load the running queue and the pending queues. This is intended to be used when you want to exit ComfyUI with queues still remaining.

Seamless video and audio continuation extension nodes (Video Extender, Video Merge, Loop Trim) for MiniMax H3 in ComfyUI.

NODES: OverrideLoadedDiffusionDevice. I happen to have two graphics cards, and I want to load models into another graphics card in Comfyui, so I designed this small node.

Various custom nodes for improved usability.

Prompt management extension providing a prompt toggle grid, global archives, tag autocomplete, localization, and a desktop workflow bridge for ComfyUI.

ComfyUI nodes that improve SDXL prompt adherence and any-size/tiled VAE workflows.

Model-aware sigma schedules, sampler compatibility validation, and versioned inference profiles for ComfyUI.

ComfyUI custom node for KittenTTS - Ultra-lightweight text-to-speech (CUDA / CPU)

SDFXBridgeForComfyUI is a custom node designed for seamless integration between ComfyUI and SDFX. This custom node allows users to make ComfyUI compatible with SDFX when running the ComfyUI instance on their local machines.

Virtual camera inside a 360 image. Real focal lengths, sensor sizes, lens projections, interactive drag framing.

A custom node that provides enhanced control over style transfer balance when using FLUX style models in ComfyUI. This node offers better control over the influence of text prompts versus style reference images.

Custom nodes for ComfyUI to generate empty latent space compatible with Hunyuan models for both image and video generation.

Highly customizable Scheduler for ComfyUI.

ComfyUI nodes for palette-card image conditioning and structured palette prompts.

you can using customnet in comfyUI

you can using pic2story in comfyUI

Generate parametric 3D heads by adjusting gender, race, expression, and head angle with sliders, output as color images, normal maps, depth maps, and masks for ControlNet conditioning. (Description by CC)

These nodes aid with batching image processing and maintaining input file names in output files and other quality of life nodes.

This is a set of custom nodes for ComfyUI designed for generating sequential long video clips based on the Wan 2.2 model architecture, handling continuity between segments and offering advanced control over the sampling process

Dynamic Position Extrapolation for Ultra High Resolution FLUX Image Generation This custom node enables FLUX models in ComfyUI to generate ultra-high resolution images (4K and beyond) using DyPE (Dynamic Position Extrapolation) technology.

Run Python code, Outline, List, Four-quadrant grid, Nine-square grid

Utility nodes for ComfyUI: logic switches, bundle nodes, selectors, bypass nodes, image/latent utilities, and workflow automation nodes.

The Joy Caption Alpha Two node for ComfyUI, based on https://huggingface.co/spaces/fancyfeast/joy-caption-alpha-two

A ComfyUI custom node for generating high-fidelity, synchronized foley audio for any video, powered by Tencent’s HunyuanVideo-Foley model.

FaithC mesh processing wrapper providing nodes for encoding, decoding, exporting, loading, normalizing, and simplifying 3D mesh models in ComfyUI. (Description by CC)