2026-02-14 12:34:12 +07:00
2025-08-09 13:22:24 +08:00
2025-09-11 17:31:49 +08:00
2025-09-11 17:31:49 +08:00
2025-04-01 15:01:19 +07:00
2025-08-09 17:28:50 +08:00
2025-09-11 17:31:49 +08:00
2025-04-01 15:01:19 +07:00
2026-02-14 12:34:12 +07:00

Civicomfy - Civitai Model Downloader for ComfyUI

Civicomfy seamlessly integrates Civitai's vast model repository directly into ComfyUI, allowing you to search, download, and organize AI models without leaving your workflow.

Features

  • Integrated Model Search: Search Civitai's extensive library directly from ComfyUI
  • One-Click Downloads: Download models with associated metadata and thumbnails
  • Automatic Organization: Models are automatically saved to their appropriate directories
  • Clean UI: Clean, intuitive interface that complements ComfyUI's aesthetic

Installation

Git clone

cd ComfyUI/custom_nodes
git clone https://github.com/MoonGoblinDev/Civicomfy.git

Comfy-CLI

comfy node registry-install civicomfy

ComfyUI Manager

Screenshot 2025-04-08 at 11 42 46

Usage

  1. Start ComfyUI with Civicomfy installed
  2. Access the Civicomfy panel from the Civicomfy menu button at the right top area.
  3. Search for models
  4. Click the download button on any model to save it to your local installation
  5. Models become immediately available in ComfyUI nodes

Configuration

  • Enter your Civitai API Token in the settings, or set CIVITAI_API_KEY in the server environment (useful for cloud deployments like RunPod).
  • Optional: set a Global Download Root in Civicomfy settings.
    • When set, Civicomfy saves to <global_root>/<model_type> (for example /runpod-volume/ComfyUI/checkpoints or /runpod-volume/ComfyUI/loras).
    • When empty, Civicomfy uses the default ComfyUI paths (folder_paths / extra_model_paths.yaml).
    • The global root is persisted on disk in custom_nodes/Civicomfy/root_settings.json.

Screenshots

Screenshot 2025-04-08 at 11 24 40 Screenshot 2025-04-08 at 11 23 17 Screenshot 2025-04-08 at 11 25 15 Screenshot 2025-04-08 at 11 25 24

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

S
Description
No description provided
Readme MIT
8.5 MiB
Languages
Python 62.8%
JavaScript 31.1%
CSS 6.1%