AI Model Wiki

    Skip to main content
    8 min read
    Lesson 2 of 6

    Setup: Automatic1111 & ComfyUI

    From the course Stable Diffusion Mastery: Local AI Image Generation

    There are two main interfaces for running Stable Diffusion locally: Automatic1111 (A1111) and ComfyUI. Each has different strengths, and many users end up using both.

    Automatic1111 (AUTOMATIC1111/stable-diffusion-webui)

    The most popular web UI for Stable Diffusion. It provides a straightforward, form-based interface.

    Installation (Windows):

    1. Install Python 3.10.6 (check "Add to PATH")
    2. Install Git
    3. Open Command Prompt and run:
    4. First launch downloads required models (several GB)
    5. Access via browser at http://127.0.0.1:7860

    Key Features:

    • Simple text-to-image and image-to-image interface
    • Built-in extension manager for adding features
    • Inpainting with mask painting directly in the browser
    • Batch generation with X/Y/Z plot comparisons
    • ControlNet integration via extension
    • Wide community support and tutorials

    ComfyUI

    A node-based workflow editor — think Blender's node system for AI images. More complex but infinitely more powerful.

    Installation:

    1. Download the latest release from GitHub
    2. Extract and run the appropriate script for your OS
    3. Install the ComfyUI Manager extension for easy model management
    4. Access via browser at http://127.0.0.1:8188

    Key Features:

    • Visual node-based workflow — you see exactly how the pipeline works
    • Reusable workflows that can be shared and loaded
    • More efficient VRAM usage than A1111
    • Advanced scheduling, multiple samplers in one generation
    • Better support for SDXL and FLUX models
    • Workflow marketplace — download complex pipelines others have created

    Which Should You Choose?

    • Beginners → Start with A1111. Simpler interface, easier to learn.
    • Advanced users → ComfyUI for complex workflows and better performance.
    • Both → Many users run both. A1111 for quick generations, ComfyUI for complex pipelines.

    Installing Models

    Models go in the models/Stable-diffusion folder (A1111) or models/checkpoints folder (ComfyUI):

    • Download models from CivitAI or Hugging Face
    • Popular checkpoints: Realistic Vision, DreamShaper, Juggernaut XL, FLUX.1 Dev
    • File sizes range from 2GB to 7GB per model

    Essential Extensions (A1111)

    • ControlNet — Guide generation with poses, edges, depth maps
    • ADetailer — Automatic face/hand detail improvement
    • Regional Prompter — Different prompts for different image regions
    • Ultimate SD Upscale — Tile-based upscaling for large images

    Start with A1111 to learn the fundamentals, then transition to ComfyUI when you want more control. The skills transfer directly.

    Key Takeaways

    • •Automatic1111 is a form-based UI — best for beginners and quick generations
    • •ComfyUI is a node-based workflow editor — more powerful and efficient for advanced use
    • •Models can be downloaded from CivitAI and Hugging Face, stored in specific folders
    • •Essential A1111 extensions: ControlNet, ADetailer, Regional Prompter
    • •Start with A1111, then learn ComfyUI for complex workflows
    0 of 6 lessons marked complete

    Progress is kept in this browser only.