WHAT IS COMFYUI?

Aug 14, 2025

CompareAI.AI

Everything you need to know about ComfyUI in simple words. What it is, how it works, and how you can use it with open-source AI models to create images and videos.

ComfyUI is a desktop application that lets you build workflows for open-source AI models. You drag and drop small boxes called nodes, connect them like a flowchart, and run the workflow to generate images or videos. You don’t have to write code. ComfyUI is popular because it is visual, fast, and gives full control over each step.

Example of workflow of ComfyUI software for creating AI images and Videos utilizing user's system resources.

This is an example of a simple comfy UI workflow to create AI images with QWEN AI model and respective Clip and VAE required to generate an AI image.

Key Features of ComfyUI

  • Visual, drag-and-drop editor
    Build image and video generation workflows by connecting nodes. Easy to understand at a glance.

  • Works with many open-source AI models
    Use image models and video models. You decide which model to load in your workflow.

  • No coding required
    Beginners can start quickly. Advanced users can still fine-tune every parameter.

  • Reusable and shareable
    Save workflows as JSON files. Import/export with one click and share with others.

  • Extensible with custom nodes
    Add community node packs to unlock new loaders, samplers, video tools, upscalers, and more.

  • Runs locally on your PC
    Keep your files private. Best experience with a modern NVIDIA GPU; CPU can work but is slow.


Setting Up ComfyUI

  1. Get ComfyUI
    The easiest way on Windows is the ComfyUI Windows installer (.exe) (Download here). It sets up everything for you.
    On Mac and Linux, you can use the manual install (Python) or community builds.

  2. Install
    Run the installer (Windows) or follow the simple steps for your OS. Choose where to keep the ComfyUI folder.

  3. Launch
    Start ComfyUI like any app. A small console window will open, and then the ComfyUI page will open in your browser.

  4. Basic Tour
    The big empty area is the canvas. Right-click or press Search to add nodes. Save your workflow from the top bar.

    Tip: ComfyUI also has built-in default templates. You can browse and load these templates directly. If any models or custom nodes are missing, ComfyUI will automatically prompt you to download them — simply click the download button to install.


Using AI Models (Image & Video)

Goal: Load the model files you want and make them available inside your workflow.

Step 1: Download models
Get models from trusted places (official model pages, Hugging Face, community hubs). Pick the model type you need: image or video.

Step 2: Put files in the Models folder
Place model files inside the ComfyUI/models directory. Many node packs create their own sub-folders (for example: checkpoints, vae, loras, control, video). If a node asks for a specific folder, use that folder.

Step 3: Refresh in ComfyUI
Open your model-loading node and click Refresh so ComfyUI sees the new files.

Step 4: Select and generate
Choose the model in the node, connect a simple workflow (for images: Load Model → Sampler → Save Image; for videos: use the video nodes provided by the node pack), set your prompt and settings, then Queue Prompt to run.

Extra Tip: You can also download custom workflows (JSON files) from the internet and open them directly in ComfyUI. If any required models or nodes are missing, ComfyUI will prompt you to download them automatically.

For example: download QWEN AI simple workflow and its resources from here.

Check out our comparisons on image and video generative AI tools like Leonardo AI, Ideogram etc. with results and analysis over each AI tool, Or just judge them yourself by comparing the results of each ai tool, go to home now!.

BLOGS

WHAT IS COMFYUI?

Aug 14, 2025

CompareAI.AI

Everything you need to know about ComfyUI in simple words. What it is, how it works, and how you can use it with open-source AI models to create images and videos.

ComfyUI is a desktop application that lets you build workflows for open-source AI models. You drag and drop small boxes called nodes, connect them like a flowchart, and run the workflow to generate images or videos. You don’t have to write code. ComfyUI is popular because it is visual, fast, and gives full control over each step.

Example of workflow of ComfyUI software for creating AI images and Videos utilizing user's system resources.

This is an example of a simple comfy UI workflow to create AI images with QWEN AI model and respective Clip and VAE required to generate an AI image.

Key Features of ComfyUI

  • Visual, drag-and-drop editor
    Build image and video generation workflows by connecting nodes. Easy to understand at a glance.

  • Works with many open-source AI models
    Use image models and video models. You decide which model to load in your workflow.

  • No coding required
    Beginners can start quickly. Advanced users can still fine-tune every parameter.

  • Reusable and shareable
    Save workflows as JSON files. Import/export with one click and share with others.

  • Extensible with custom nodes
    Add community node packs to unlock new loaders, samplers, video tools, upscalers, and more.

  • Runs locally on your PC
    Keep your files private. Best experience with a modern NVIDIA GPU; CPU can work but is slow.


Setting Up ComfyUI

  1. Get ComfyUI
    The easiest way on Windows is the ComfyUI Windows installer (.exe) (Download here). It sets up everything for you.
    On Mac and Linux, you can use the manual install (Python) or community builds.

  2. Install
    Run the installer (Windows) or follow the simple steps for your OS. Choose where to keep the ComfyUI folder.

  3. Launch
    Start ComfyUI like any app. A small console window will open, and then the ComfyUI page will open in your browser.

  4. Basic Tour
    The big empty area is the canvas. Right-click or press Search to add nodes. Save your workflow from the top bar.

    Tip: ComfyUI also has built-in default templates. You can browse and load these templates directly. If any models or custom nodes are missing, ComfyUI will automatically prompt you to download them — simply click the download button to install.


Using AI Models (Image & Video)

Goal: Load the model files you want and make them available inside your workflow.

Step 1: Download models
Get models from trusted places (official model pages, Hugging Face, community hubs). Pick the model type you need: image or video.

Step 2: Put files in the Models folder
Place model files inside the ComfyUI/models directory. Many node packs create their own sub-folders (for example: checkpoints, vae, loras, control, video). If a node asks for a specific folder, use that folder.

Step 3: Refresh in ComfyUI
Open your model-loading node and click Refresh so ComfyUI sees the new files.

Step 4: Select and generate
Choose the model in the node, connect a simple workflow (for images: Load Model → Sampler → Save Image; for videos: use the video nodes provided by the node pack), set your prompt and settings, then Queue Prompt to run.

Extra Tip: You can also download custom workflows (JSON files) from the internet and open them directly in ComfyUI. If any required models or nodes are missing, ComfyUI will prompt you to download them automatically.

For example: download QWEN AI simple workflow and its resources from here.

Check out our comparisons on image and video generative AI tools like Leonardo AI, Ideogram etc. with results and analysis over each AI tool, Or just judge them yourself by comparing the results of each ai tool, go to home now!.

BLOGS

WHAT IS COMFYUI?

Aug 14, 2025

CompareAI.AI

Everything you need to know about ComfyUI in simple words. What it is, how it works, and how you can use it with open-source AI models to create images and videos.

ComfyUI is a desktop application that lets you build workflows for open-source AI models. You drag and drop small boxes called nodes, connect them like a flowchart, and run the workflow to generate images or videos. You don’t have to write code. ComfyUI is popular because it is visual, fast, and gives full control over each step.

Example of workflow of ComfyUI software for creating AI images and Videos utilizing user's system resources.

This is an example of a simple comfy UI workflow to create AI images with QWEN AI model and respective Clip and VAE required to generate an AI image.

Key Features of ComfyUI

  • Visual, drag-and-drop editor
    Build image and video generation workflows by connecting nodes. Easy to understand at a glance.

  • Works with many open-source AI models
    Use image models and video models. You decide which model to load in your workflow.

  • No coding required
    Beginners can start quickly. Advanced users can still fine-tune every parameter.

  • Reusable and shareable
    Save workflows as JSON files. Import/export with one click and share with others.

  • Extensible with custom nodes
    Add community node packs to unlock new loaders, samplers, video tools, upscalers, and more.

  • Runs locally on your PC
    Keep your files private. Best experience with a modern NVIDIA GPU; CPU can work but is slow.


Setting Up ComfyUI

  1. Get ComfyUI
    The easiest way on Windows is the ComfyUI Windows installer (.exe) (Download here). It sets up everything for you.
    On Mac and Linux, you can use the manual install (Python) or community builds.

  2. Install
    Run the installer (Windows) or follow the simple steps for your OS. Choose where to keep the ComfyUI folder.

  3. Launch
    Start ComfyUI like any app. A small console window will open, and then the ComfyUI page will open in your browser.

  4. Basic Tour
    The big empty area is the canvas. Right-click or press Search to add nodes. Save your workflow from the top bar.

    Tip: ComfyUI also has built-in default templates. You can browse and load these templates directly. If any models or custom nodes are missing, ComfyUI will automatically prompt you to download them — simply click the download button to install.


Using AI Models (Image & Video)

Goal: Load the model files you want and make them available inside your workflow.

Step 1: Download models
Get models from trusted places (official model pages, Hugging Face, community hubs). Pick the model type you need: image or video.

Step 2: Put files in the Models folder
Place model files inside the ComfyUI/models directory. Many node packs create their own sub-folders (for example: checkpoints, vae, loras, control, video). If a node asks for a specific folder, use that folder.

Step 3: Refresh in ComfyUI
Open your model-loading node and click Refresh so ComfyUI sees the new files.

Step 4: Select and generate
Choose the model in the node, connect a simple workflow (for images: Load Model → Sampler → Save Image; for videos: use the video nodes provided by the node pack), set your prompt and settings, then Queue Prompt to run.

Extra Tip: You can also download custom workflows (JSON files) from the internet and open them directly in ComfyUI. If any required models or nodes are missing, ComfyUI will prompt you to download them automatically.

For example: download QWEN AI simple workflow and its resources from here.

Check out our comparisons on image and video generative AI tools like Leonardo AI, Ideogram etc. with results and analysis over each AI tool, Or just judge them yourself by comparing the results of each ai tool, go to home now!.

BLOGS