ComfyUI Custom Nodes
Discover and install open-source custom node extensions for ComfyUI. Click any repository to view code and installation scripts.
No nodes match your filter
Try selecting "All" or clearing your search term.
Audio Separation Nodes
Audio source separation node suite employing Demucs models to split vocal, drum, bass, and instrumental stems for video audio sync.
Audio vocal/drum stem separation via Demucs for audio-reactive AI video
Use Everywhere
Wireless connection nodes automatically broadcasting inputs across matching data types without cluttering canvas wires.
Wireless node connections and clean workflow canvas organization
ComfyUI CLIPtion
Automated image captioning and prompt interrogator node utilizing CLIP and BLIP models for dataset tagging.
Automated image captioning & dataset tagging via CLIP/BLIP
MTB Nodes
Multi-purpose utility nodes featuring face swap helpers, animation interpolation tools, color grading, and latent masks.
Color grading, face swap pipelines, and latent mask editing
Comfy WaveSpeed
First-block caching and wave speed acceleration optimizations for FLUX.1 and diffusion inference speedups.
Accelerating FLUX.1 & diffusion sampling speed without quality loss
Comfyroll Studio
Versatile suite of design, animation, aspect ratio, text formatting, and workflow control nodes.
Aspect ratio control, animation utilities, and layout formatting
ComfyUI 3D Pack
Comprehensive 3D processing suite supporting Mesh reconstruction, UV mapping, 3D Gaussian Splatting (3DGS), and NeRF rendering.
3D mesh generation, Gaussian Splatting, and NeRF rendering inside ComfyUI
Advanced ControlNet
Advanced scheduling, weights control, multi-ControlNet stacking, and custom timestep control for diffusion models.
Advanced ControlNet scheduling, weight decay, and multi-model stacking
Advanced LivePortrait
Advanced facial expression driving and motion transfer nodes for real-time and high-precision portrait animation.
Real-time portrait motion transfer & facial expression driving
AnimateDiff Evolved
Advanced motion module loading, sliding window inference, and prompt travel control for AI video and animation generation.
Fluid AI video generation & camera motion control
Automatic CFG
Dynamic CFG scaling node automatically adjusting guidance scale per sampling step to eliminate saturation and boost detail.
Automatic CFG step scaling to eliminate burn & increase clarity
ComfyUI ControlNet Aux
Preprocessors pack containing DepthAnything V2, DWPose, Canny, Lineart, OpenPose, and HED boundary extractors.
Extracting pose, depth maps, line art, and structural edges from input photos
ComfyUI Copilot
AI assistant node integration helping automate workflow generation, node wiring, and prompt suggestions inside ComfyUI.
AI copilot workflow generation, auto-wiring, and prompt assistance
AlekPet Custom Nodes
Custom nodes collection featuring Google Translate prompt translation, DeepL integration, painter node canvas, and Pose Maker.
Prompt translation (DeepL/Google), canvas painter, and Pose Maker
ComfyUI Custom Scripts
UI enhancement scripts and utility nodes featuring auto-complete Lora/Embedding prompts, image previewers, and math expressions.
Prompt auto-complete, canvas enhancements, and custom UI helpers
Deno Custom Nodes
NVIDIA RTX Video Super Resolution, NVIDIA VFX filters, and acceleration helpers for LTX and Wan video models.
NVIDIA RTX Video Super Resolution & NVIDIA VFX filters
ComfyUI-Easy-Use
Workflow simplification suite providing easy sampling pipes, wildcards, one-click upscaling, and user-friendly node presets.
Workflow simplification, easy sampler pipes, and wildcard prompts
ComfyUI Essentials
Core Quality-of-Life nodes collection by Cubiq for image manipulation, mask operations, math expressions, and data routing.
Essential image/mask operations, math calculations, and QoL helpers
ComfyUI fal.ai API
Cloud inference bridge to fal.ai API providing ultra-fast cloud GPU generation for image, video, audio, and LLM models.
1-click cloud GPU inference via fal.ai API for FLUX & video models
Frame Interpolation (VFI)
Video frame interpolation suite incorporating RIFE, FILM, AMT, and FLAVR for ultra-smooth slow-motion video generation.
RIFE & FILM video frame interpolation and smooth slow-motion video
ComfyUI GGUF
GGUF quantized model loader for FLUX.1 and SDXL models, significantly reducing VRAM usage for consumer GPUs.
Running FLUX.1 & SDXL models on low-VRAM GPUs with GGUF quantizations
ComfyUI Hunyuan3DWrapper
Tencent Hunyuan3D model integration wrapper for single-image to 3D textured mesh reconstruction.
Image-to-3D textured mesh reconstruction via Hunyuan3D
ComfyUI HunyuanVideoWrapper
Dedicated integration wrapper for Tencent HunyuanVideo open-source video diffusion model text-to-video workflows.
Tencent HunyuanVideo high-definition text-to-video generation
ComfyUI IC-Light
Illumination Connection (IC-Light) nodes for consistent subject re-lighting, background light blending, and normal map relighting.
Consistent subject re-lighting and background light harmony
ComfyUI Impact Pack
FaceDetailer pipe automation, SAM segmentation, detector nodes, and iterative upscaling workflow tools.
Automated face repair, SAM segmentation, and detailer pipes
Impact Subpack
Companion detector module for Impact Pack integrating Ultralytics YOLO models and custom object detection weights.
Ultralytics YOLO detector weights for Impact Pack FaceDetailer
ComfyUI Inpaint Nodes
Advanced inpainting node suite integrating Fooocus, LaMa, MAT, and outpainting algorithms for seamless image editing.
Fooocus inpainting, LaMa object removal, and seam-free outpainting
ComfyUI Inspire Pack
Companion extension for Impact Pack adding Regional Prompting, Variation Seed sliders, and Global Sampler controls.
Regional Prompting, Variation Seeds, and Global Sampler control
ComfyUI InstantID
Zero-shot identity-preserving generation using single facial photo inputs with strong ControlNet keypoint guidance.
Zero-shot character identity preservation & face generation
IPAdapter Plus
Image-prompt adapter nodes enabling reference image style transfer, face cloning, and image prompts without LoRA training.
Zero-shot style transfer & character face consistency
ComfyUI Janus-Pro
DeepSeek Janus-Pro multimodal vision-language model integration for image understanding, visual QA, and prompt generation.
DeepSeek Janus-Pro multimodal vision understanding & image analysis
ComfyUI-KJNodes
Extensive utility node suite providing image transformations, Set/Get variable nodes, color masks, and hidden wire organization.
Set/Get variable routing, image transformations, and canvas wire organization
ComfyUI LayerDiffusion
Transparent image generation and layer diffusion nodes producing native PNG images with alpha channel transparency.
Transparent PNG image generation & alpha channel layer diffusion
ComfyUI LayerStyle
Photoshop-style layer compositing, blending modes, advanced mask extraction, and background removal tools.
Photoshop-style layer compositing, masking, and background removal
ComfyUI LayerStyle Advance
Advanced Photoshop layer style extension adding procedural layer effects, drop shadows, inner glows, and text compositing.
Advanced layer effects, drop shadows, inner glows, and text styling
ComfyUI LTX-Video
Lightricks LTX-Video model integration and LoRA loader nodes for real-time high-definition video generation.
Lightricks LTX-Video model loading & motion LoRAs
ComfyUI Lux3D
Image-to-3D generation, mesh reconstruction, and real-time viewport visualization suite.
Image-to-3D generation & interactive 3D viewport visualization
ComfyUI-Manager
The core extension manager for ComfyUI enabling one-click node installation, missing node auto-detection, and model downloads.
1-click node installation, updates, and missing node fixes
Mixlab Nodes
Comprehensive workflow suite featuring Web App UI builder, screen canvas, speech-to-text, and prompt translation.
Web App UI builder, screen canvas drawing, and speech-to-text
ComfyUI MMAudio
Multimodal video-to-audio synthesis engine generating synchronized sound effects and background audio for video inputs.
Generating synchronized sound effects & audio for AI videos
ComfyUI MultiGPU
Multi-GPU execution and device load-balancing optimization for parallel inference across multi-card systems.
Multi-GPU parallel processing & VRAM load balancing
ComfyUI MVAdapter
Multi-View consistent image generation adapter producing 360-degree object views and consistent camera angles.
Multi-view 360-degree consistent object and character generation
ComfyUI N-Sidebar
Interactive UI sidebar enhancement introducing node drag-and-drop, favorite node bookmarks, and instant category search.
UI sidebar node bookmarks, drag-and-drop, and category search
ComfyUI Nunchaku
SVDQuant 4-bit quantization acceleration engine enabling ultra-fast FLUX.1 inference with minimal VRAM footprint.
4-bit SVDQuant acceleration for fast FLUX.1 generation
ComfyUI Patches (LL)
Performance optimization hooks providing TeaCache acceleration, First Block Cache, and VRAM reduction patches.
TeaCache acceleration hooks & First Block Cache optimization
Portrait Master
Comprehensive portrait prompt generator and parameter tuning node for skin details, lighting, and camera angles.
Generating detailed portrait prompts, facial features, and lighting
Prompt Auto Complete
Real-time prompt completion and tag suggestion node for Danbooru tags, embeddings, and LoRA triggers.
Danbooru tag auto-complete, LoRA trigger search, and prompt suggestions
ComfyUI PuLID Flux
Pure and Lightning Identity (PuLID) tuning node specifically engineered for high-fidelity facial identity preservation in FLUX.1.
FLUX.1 zero-shot facial identity preservation
ComfyUI PuLID Flux (LL)
Enhanced PuLID Flux implementation with optimized facial landmark extraction and multi-face conditioning weights.
Facial landmark extraction & multi-face PuLID conditioning
ReActor Face Swap
Fast and simple face swapping node extension utilizing InsightFace for seamless single and multi-face swapping.
Fast 1-click single & multi-person face swapping
ComfyUI RMBG
High-precision background removal nodes utilizing BRIA AI RMBG models for instant product cutout generation.
Instant 1-click background removal & product cutout extraction
SeedVR2 Video Upscaler
Dedicated SeedVR2 video upscaling wrapper for temporal video frame interpolation and high-resolution video enhancements.
High-resolution AI video upscaling and temporal frame enhancement
Segment Anything 2 (SAM2)
Meta SAM2 integration for real-time video object tracking, high-precision image segmentation, and mask generation.
SAM2 video object tracking and high-precision mask extraction
Spectrum MiniMax-H3
Dedicated spectral acceleration module for MiniMax H3 video models delivering up to 30% faster inference speeds.
30% faster inference for MiniMax H3 video models via spectral acceleration
ComfyUI StarNodes
Emerging comprehensive toolkit containing wildcards, PSD layer import, image processing filters, and workflow tools.
PSD file import, wildcard prompt expansion, and image filters
ComfyUI SUPIR
Photorealistic image restoration and high-fidelity super-resolution upscaling wrapper based on the SUPIR model architecture.
Ultra-high quality photo restoration, face enhancement, and super-resolution
ComfyUI TouchDesigner Bridge
Real-time streaming integration bridge connecting ComfyUI image generation pipelines directly with TouchDesigner for live AV shows.
Real-time streaming between ComfyUI and TouchDesigner for live visual shows
ComfyUI TeaCache
Timestep embedding caching optimization for FLUX.1 and LTX-Video models accelerating sampling speeds up to 2x without quality drop.
2x sampling acceleration for FLUX.1 & LTX-Video via timestep caching
Ultimate SD Upscale (GGUF)
Enhanced Ultimate SD Upscale fork with GGUF model support for seamless 4K/8K tile upscaling on low-VRAM GPUs.
4K/8K tile upscaling with GGUF quantized models
Video Frame Interpolation Advanced
Advanced video frame interpolation module supporting optical flow estimation, RIFE 4.26, and custom framerate multipliers.
Optical flow video frame interpolation & custom framerate multipliers
Video Helper Suite (VHS)
Essential video loading, frame combining, MP4/GIF exporting, and audio-video synchronization node suite.
Loading video files, exporting MP4/GIF, and frame sequence processing
Wan Start-End Frames Native
Native start and end frame latent conditioning wrapper for Wan2.1 video models providing smooth temporal interpolation.
Native start/end frame latent interpolation for Wan2.1
WanVideo Start-End Frames
First and last keyframe conditioning control node for Wan video diffusion models enabling seamless video looping and keyframe transitions.
Wan video first & last frame keyframe control & video looping
ComfyUI WanVideoWrapper
Dedicated integration wrapper for Wan video diffusion models enabling high-quality text-to-video and image-to-video generation.
Wan video model generation, text-to-video, and image-to-video workflows
Z-Image Power Nodes
Specialized power node toolkit optimized for next-gen image models with advanced latent conditioning and pipeline helpers.
Z-Image model latent conditioning & specialized power pipeline tools
rgthree-comfy
Workflow optimization suite providing Context bus nodes, Fast Mute/Bypass toggles, Power Prompting, and seed controllers.
Clean workflow routing, context busing, and fast mute/bypass controls
Dynamic Thresholding (dwt)
Prevents color burn and oversaturation at high CFG scales by dynamically clamping latent values during sampling.
High CFG sampling without image burn or oversaturation
TimeSlice Motion Graphics
Motion graphics and post-processing nodes providing time-slice effects, slit-scan animation, and custom video FX.
Time-slice motion graphics, slit-scan video FX, and post-processing
WAS Node Suite
Comprehensive Swiss-army-knife node collection featuring hundreds of image processing, text manipulation, and mathematical utilities.
Swiss-army-knife image editing, text manipulation, and math utilities
LTX Director (WhatDreamsCost)
Timeline animation editor for LTX-Video enabling multi-prompt keyframing, audio reactive controls, and shot transitions.
LTX-Video timeline keyframing, multi-prompt shots, and audio reactive controls
Showing 1–24 of 70 nodes