Comfy with ComfyUI
  • How to Use This Site
  • Get It Running
    • What is ComfyUI?
    • Recommended Specs
    • Setup
    • Comfy Cloud
    • First Run
  • PC Basics
    • Path
    • Terminal
    • Python
    • git
  • Customization
    • Settings
    • Command Line Arguments
    • Updates
    • Downloading & Placing Models
    • Custom Nodes
  • ComfyUI Operations
    • Run & Stop
    • Save & Load Workflows
    • Canvas Navigation
    • Nodes
    • Group
    • Media
    • Mask Editor
    • Subgraphs
  • Columns
    • Readable Nodes
    • Wireless Nodes
  • Data Operations
    • Data Types
    • Simple Math
    • Text Operations
    • Webcam Input
  • Basic Image Processing
    • Resize, Crop, Pad
    • Color Adjustments & Filters
    • Mask & Alpha Channel
    • Mask Operations
    • AI Mask Generation
    • Layer Composite & Blend
  • Batch Processing
    • Queue
    • List
    • Batch & Video
    • Realtime Processing
  • Mechanics of Image Generation AI
    • Diffusion Models
    • Conditioning
    • CFG
    • Sampling
    • Latent Diffusion Model & VAE
  • Applications of Image Generation
    • Upscale & Restoration
    • ControlNet Family
    • Region-Limited Generation
    • Subject Transfer
    • ID Transfer & FaceSwap
    • Virtual Try-On
    • Style Transfer
    • Line Art Coloring
    • Object Removal
    • Instruction-Based Image Editing
    • Refining Rough Collages
  • Image Translation
    • Relight
    • Depth Estimation & Normal Map
  • Video & Audio Generation
    • Video Generation
    • Video Upscale & Restoration
    • Frame Interpolation
    • video2audio
    • Lip Sync
    • talking head
    • Human Motion Transfer
    • TTS
    • Voice Clone
    • Music Generation
  • 3D Generation
    • 3D Model Generation
  • Computer Vision / VLM
    • Object Detection
    • Matting
    • Segmentation
  • LLM / MLLM Usage
    • Prompt Generation & Editing
    • Tag & Caption Generation
  • SD1.5 Basics
    • text2image
    • Textual Inversion
    • LoRA
    • KSampler (Advanced) Node
    • image2image
    • Hires.fix
    • inpainting
    • outpainting
    • Differential Diffusion
    • ControlNet
      • ControlNet Preprocessor
    • IP-Adapter
  • Other Foundation Models
    • SDXL
      • SDXL Anime
    • Flux.1
      • Flux.1 Tools
      • ACE++
      • Flux.1 Kontext
      • Chroma1-HD
    • FLUX.2 [klein]
    • AuraFlow
    • Lumina-Image 2.0
    • Qwen-Image
      • Qwen-Image-Edit
      • Qwen-Image-Layered
    • Z-Image-Turbo
  • Upscale & Fix
    • ESRGAN
    • Ultimate SD Upscale
    • Detailer
  • Video Workflows
    • Wan 2.1
      • Wan 2.1 VACE
      • Wan-Animate
    • Wan 2.2
    • LTX-2
    • FramePack
      • FramePack One Frame Inference
    • LivePortrait
  • LLM / MLLM
    • JoyCaption
    • Florence-2
    • Qwen 3 VL
  • Other
    • ReActor
  • System Optimization
    • Model Merge & Difference LoRA
    • Speed and Efficiency
  • API Nodes
  • What to do when an error occurs
  • Common Questions
    • Why generate at 512px × 512px?
    • Can't use SD1.5 LoRA / ControlNet with SDXL?
    • Why can only resolutions that are multiples of 8 be generated?
    • Can't generate the same image with web UI and ComfyUI?
  • Bad Output
    • People or objects are duplicating
  • Error Codes
    • (IMPORT FAILED)
    • torch.cuda.OutOfMemoryError (VRAM shortage)
About
日本語 English

News

2026.1.18 basic-workflows Added FLUX.2 [klein] page 2026.1.18 data-utilities Added description for Resize Image/Mask node 2026.1.17 basic-workflows Updated LTX-2 page
Adjusted parameters and added many workflows
2026.1.15 none Created the updates page
Comfy with ComfyUI Docs
How to use this site About News GitHub

What is the JSON copy button?

It exports every node, connection, and parameter so someone else can reproduce the exact workflow.

Select the Copy JSON button on any workflow article to place the graph on your clipboard, then press Ctrl+V inside ComfyUI to load it.

This page has an issue!

Include a short repro step or the correct information if you can. We attach the current URL automatically.

Double-check before sending. We'll create an anonymous GitHub issue.

We'll file a GitHub issue—please avoid personal data.

Please explain more!

Tell us what topic or model you need so we can plan the write-up.

Double-check before sending. We'll create an anonymous GitHub issue.

We'll file a GitHub issue—please avoid personal data.

Thank you

We’ll use your feedback to keep improving.

Category