ComfyUI Extension: ComfyUI Ethnic Outfits Custom Nodes
Custom ComfyUI nodes for generating outfit prompts representing diverse ethnicities, cultures, and regions. Uses extensible JSON data for clothing, accessories, and poses, with โrandom/disabledโ dropdowns for flexibility. Advanced prompt engineering is supported via Ollama LLM integration. Easily add new regions, ethnicities, or cultures by updating data files and creating lightweight node wrappers. Designed for artists, researchers, and developers seeking culturally rich, customizable prompt generation in ComfyUI workflows.
Custom Nodes (0)
README
ComfyUI Ethnic Outfits Custom Nodes
Custom ComfyUI nodes for generating outfit prompts representing diverse ethnicities, cultures, and regions. Uses extensible JSON data for clothing, accessories, and poses, with โrandom/disabledโ dropdowns for flexibility. Advanced prompt engineering is supported via Ollama LLM integration. Easily add new regions, ethnicities, or cultures by updating data files and creating lightweight node wrappers. Designed for artists, researchers, and developers seeking culturally rich, customizable prompt generation in ComfyUI workflows.
Features
- Unified codebase for multiple regions, ethnicities, and cultures
- Shared, extensible JSON data for clothing, accessories, and poses
- Consistent dropdown options (with "random" and "disabled" support)
- Easily add new regions, ethnicities, cultures, or customize logic
- Advanced prompt engineering for SDXL/Flux/LLM nodes
Supported Regions
- Example: India ๐ฎ๐ณ, Indonesia ๐ฎ๐ฉ, Japan ๐ฏ๐ต (add your own!)
Directory Structure
custom_nodes/comfyui-outfits/
โโโ data/
โ โโโ common/
โ โโโ styles/
โ โโโ prompts/
โ โโโ in/
โ โโโ id/
โ โโโ jp/
โ โโโ ...
โโโ nodes/
โ โโโ ethnic_outfit_common.py
โ โโโ in_node.py
โ โโโ id_node.py
โ โโโ jp_node.py
โ โโโ ollama_prompter.py
โ โโโ ...
โโโ __init__.py
โโโ README.md
โโโ ...
Installation (Custom Node)
- Download or clone this repository:
git clone https://github.com/manifestations/comfyui-outfits.git
- Copy or move the
comfyui-outfits
folder into your ComfyUIcustom_nodes
directory. For example:- On Windows:
C:\Users\<yourname>\ComfyUI\custom_nodes\comfyui-outfits
- On Linux/macOS:
~/ComfyUI/custom_nodes/comfyui-outfits
- On Windows:
- Restart ComfyUI.
- The new nodes will appear in the ComfyUI node list under "๐WizDroid/PromptGen".
Usage
- Use the "Ethnic Outfit Generator" nodes in your workflow.
Adding New Regions
- Add new data files under
data/<region_code>/
. - Create a lightweight node wrapper (e.g.,
xx_node.py
) using the shared class. - Register the node in
__init__.py
.
Ollama Prompter & LLM Integration
This extension includes an Ollama Prompter node for advanced prompt engineering with SDXL, Flux, and LLMs.
Ollama Installation
- Visit Ollama's official site and download the installer for your OS.
- Follow the installation instructions for your platform (Windows, macOS, Linux).
- Start the Ollama server:
- On Windows: Run
ollama serve
in Command Prompt or PowerShell. - On macOS/Linux: Run
ollama serve
in Terminal.
- On Windows: Run
Llama 3.2 Model Installation
- After installing Ollama, pull the Llama 3.2 model:
(Replaceollama pull llama3:latest
llama3:latest
with the specific tag if needed.) - Verify the model is available:
ollama list
- The Ollama Prompter node will now be able to use Llama 3.2 for prompt enhancement.
Using the Ollama Prompter Node
- Add the "Ollama Prompter" node to your ComfyUI workflow.
- Configure the prompt and model options as needed.
- Ensure the Ollama server is running and the desired model is available.
For more details, see the Ollama documentation and Llama 3.2 model info.
Example Workflow
Below is a screenshot of a sample ComfyUI workflow using these nodes:
License
See LICENSE for details.
Acknowledgements
Special thanks to dagthomas and rikkar69 for their inspiring work in the ComfyUI and generative AI community.
Big thanks to GitHub Copilot for AI-powered coding assistance.
Contributions and feedback welcome!