Getting Started Tutorial

Quick start guide for ComfyUI.

Introduction

ComfyUI is a powerful node-based Stable Diffusion interface. This tutorial will help you get started quickly.

First Steps

1. Start ComfyUI

Run the following command:

# Navigate to ComfyUI directory
cd ComfyUI

# Run on Windows
python main.py

# Run on Linux/Mac
python3 main.py

2. Open Browser

Open your browser and visit:

http://127.0.0.1:8188

Interface Overview

Area	Function
Left Panel	Node library
Center	Workflow canvas
Right Panel	Properties
Bottom	Generation queue

Basic Workflow

Load Checkpoint → CLIP Text Encode → KSampler → VAE Decode → Save Image

Generate Your First Image

1. Drag CheckpointLoader node
2. Load a model (e.g., v1-5-pruned-emaonly.safetensors)
3. Add CLIPTextEncode for positive prompt
4. Add CLIPTextEncode for negative prompt
5. Add KSampler
6. Add VAEDecode
7. Add SaveImage node
8. Click "Queue Prompt" to generate

Next Steps

• First Image - Generate your first image
• Prompt Basics - Write better prompts
• How to Run - More startup options