Comfyui

ComfyUI - Node-Based AI Workflow Engine

Visual node-based workflow editor for AI image generation, audio processing, and more. Connect any AI model into reusable workflows.

When to Use

• User wants to build complex AI pipelines visually
• User needs to chain multiple AI models together
• User wants reusable workflow templates
• User needs batch AI processing

Installation

git clone https://github.com/comfyanonymous/ComfyUI.git
cd ComfyUI
pip install -r requirements.txt

Quick Start

python execution/main.py --comfy-path ./

Open browser at http://localhost:8188 for the visual workflow editor.

Core Features

• Node-based visual workflow builder
• Drag and drop connections
• Batch processing
• Workflow saving/loading
• Custom node support
• Model management
• Live preview

Common Workflows

• Image generation (SD, FLUX, etc.)
• Image-to-image
• Inpainting / Outpainting
• ControlNet conditioning
• Upscaling / Enhancement
• Video generation
• Audio processing
• Voice cloning pipelines
• RVC training workflows
• Song generation chains

Node Categories

• Loaders: Load models, images, audio, video
• Model: Load and configure AI models
• Sampler: Configure generation parameters
• Image: Process and combine images
• Audio: Process audio files
• Branch: Conditional workflows
• Group: Organize nodes

Custom Nodes

pip install comfyui-manager
# Then use the built-in Custom Nodes manager in the UI

HeartMuLa ComfyUI Node

For music generation with HeartMuLa:

# Custom node by benjiyaya
# pip install HeartMuLa_ComfyUI

Related

• flux-gen: FLUX image generation
• rvc: Voice conversion
• song-generation: Music generation
• heartmula: Music generation with ComfyUI node