ComfyUI OpenAI Node

A custom ComfyUI node that integrates OpenAI-compatible APIs for prompt generation and enhancement, specifically designed for Stable Diffusion workflows.

Features

• 🤖 OpenAI Compatible API Integration: Works with any OpenAI-compatible API endpoint
• 🎨 Stable Diffusion Prompt Enhancement: Automatically expands and improves prompts for better image generation
• 🧠 DeepSeek R1 Support: Pre-configured for DeepSeek-R1-Distill-Qwen-32B-910A model
• 🧹 Clean Output: Automatically removes thinking tags (<think> blocks) from responses
• ⚙️ Flexible Configuration: Customizable model, temperature, tokens, and API settings

Installation

1. Navigate to your ComfyUI custom nodes directory:

   cd ComfyUI/custom_nodes/
   

2. Clone this repository:

   git clone https://github.com/PICOPON/ComfyUI-API-OpenAI-Node.git
   

3. Install required dependencies:

   pip install requests
   

4. Restart ComfyUI

Usage

1. Add the OpenAI Node to your ComfyUI workflow
2. Configure the node parameters according to your needs
3. Connect the output to your desired downstream nodes

Node Parameters

Parameter	Type	Default	Description
prompt	STRING	"user prompts"	The input prompt to be processed
system_prompt	STRING	Stable Diffusion prompt generator	System instructions for the AI
model	STRING	"DeepSeek-R1-Distill-Qwen-32B"	Model name to use
api_url	STRING	"http://127.0.0.1:1935/v1"	API endpoint URL
api_key	STRING	"123"	API authentication key
temperature	FLOAT	0.6	Creativity level (0.0-1.0)
max_tokens	INT	1024	Maximum response length

Default System Prompt

The node comes pre-configured with a system prompt optimized for Stable Diffusion:

You are a prompt generation AI. Your task is to take a user input for a stable diffusion prompt and output and expand the supplied prompt in a stable diffusion format to provide better output. Do not deviate from the format. Do not output anything other than a stable diffusion prompt.

Examples

Basic Usage

1. Input: "a cat"
2. Output: "a beautiful fluffy cat, detailed fur texture, bright eyes, sitting pose, soft lighting, high quality, photorealistic, 8k resolution"

Advanced Workflow

Text Input → OpenAI Node → Text Processing → Image Generation

参考流程文件AI_generated_Flux_flow.json

API Compatibility

This node works with:

• ✅ OpenAI API
• ✅ DeepSeek API
• ✅ Local LLM servers (Ollama, LM Studio, etc.)
• ✅ Any OpenAI-compatible endpoint

Configuration Examples

Local Server (Ollama, vllm, mindie .etc.)

• API URL: http://localhost:1935/v1
• Model: DeepSeek-R1-Distill-Qwen-32B or any installed model
• API Key: Leave as "123" or empty

DeepSeek API

• API URL: https://api.deepseek.com/v1
• Model: deepseek-chat
• API Key: Your DeepSeek API key

OpenAI API

• API URL: https://api.openai.com/v1
• Model: gpt-4 or gpt-3.5-turbo
• API Key: Your OpenAI API key

Features in Detail

Automatic Tag Cleaning

The node automatically removes <think> and </think> tags and their content from the AI response, ensuring clean output for downstream processing.

Error Handling

• Network timeout protection (100 seconds)
• HTTP error status handling
• Exception catching with descriptive error messages

Flexible API Integration

• Automatic URL completion for OpenAI-compatible endpoints
• Optional API key authentication
• Configurable model parameters

Troubleshooting

Common Issues

API Connection Error

• Check if your API URL is correct and accessible
• Verify your API key if authentication is required
• Ensure your local server is running (for local APIs)

Model Not Found

• Verify the model name is correct for your API provider
• Check if the model is available in your account/server

Timeout Issues

• Increase timeout value in the code if needed
• Check your internet connection
• Verify API server status

Contributing

1. Fork the repository
2. Create your feature branch (git checkout -b feature/AmazingFeature)
3. Commit your changes (git commit -m 'Add some AmazingFeature')
4. Push to the branch (git push origin feature/AmazingFeature)
5. Open a Pull Request

License

This project is licensed under the GNU-3.0 License - see the LICENSE file for details.

Acknowledgments

• ComfyUI team for the excellent framework
• OpenAI for the API standard
• DeepSeek for their powerful models

Support

If you encounter any issues or have questions:

1. Check the Issues page
2. Create a new issue with detailed information
3. Include your ComfyUI version and error logs

---

⭐ If this node helps your workflow, please consider giving it a star!