Eigen AI FLUX API Plugin

Eigen AI FLUX API Plugin for ComfyUI

A ComfyUI plugin that integrates with the Eigen AI FLUX API, providing high-quality image generation with LoRA support.

---

🚀 Quick Installation

1. Install ComfyUI

If you haven’t installed ComfyUI yet:

# Clone ComfyUI
git clone https://github.com/comfyanonymous/ComfyUI.git
cd ComfyUI

# Install dependencies
pip install -r requirements.txt

# (Optional) Start ComfyUI
python main.py

---

2. Install the Plugin

Method A: Install via ComfyUI-Manager (Recommended)

1. Install ComfyUI-Manager if not already installed:

   cd ComfyUI/custom_nodes
   git clone https://github.com/ltdrdata/ComfyUI-Manager.git
   

   Restart ComfyUI after installation.

2. Install this plugin via Manager:

   1. Start ComfyUI:

      python main.py
      

   2. Open ComfyUI-Manager

   3. Search for “Eigen AI FLUX API” or “flux api”

   4. Click Install and wait until it finishes

   5. Restart ComfyUI

   6. Verify nodes appear in the left panel by searching “Eigen AI FLUX API”

---

Method B: Manual Install

If you prefer manual installation:

cd ComfyUI/custom_nodes/
git clone https://github.com/jialuw0830/flux_api_comfyui_plugin.git

# (Optional) Start ComfyUI
cd ComfyUI
python main.py

---

3. Post-Installation Steps

1. Restart ComfyUI

2. Search for “Eigen AI FLUX API” in the node browser

3. Generated images will be saved in:

   ComfyUI/output
   

---

🎨 Key Features

• High-Quality Generation — Uses FLUX.1-schnell model
• Multi-LoRA Support — Up to 3 LoRAs simultaneously
• Upscaling — Supports 2× and 4× upscaling
• Fast Performance — Uses quantized models for quick generation

---

🔧 Node Descriptions

1. Text Node (EigenAITextNode)

Purpose: Text prompt processing and formatting Input: Text prompt Output: Processed prompt (PROMPT type) Usage:

• Add text prompt in the textarea
• Connect output to any generator node
• Supports multiline text up to 2000 characters

2. LoRA Node (EigenAILoraNode)

Purpose: LoRA model configuration and management Input: LoRA file paths and weights Output: LoRA configuration (LORA_CONFIG type) Usage:

• Set up to 3 LoRA models simultaneously
• Configure LoRA weights (0.0 to 2.0)
• Connect to generator nodes for LoRA-enhanced generation

3. Qwen Generator (EigenAIQwenGeneratorNode)

Purpose: Image generation using Qwen-compatible model Input: Prompt + LoRA config + generation parameters Output: Generated image Usage:

• Connect prompt from Text Node
• Connect LoRA config from LoRA Node
• Adjust width, height, guidance scale, and seed
• Default API: http://74.81.65.108:8010

4. Schnell Generator (EigenAISchnellGeneratorNode)

Purpose: Fast image generation using Schnell model Input: Prompt + LoRA config + generation parameters Output: Generated image Usage:

• Similar to Qwen Generator but optimized for speed
• Ideal for batch processing
• Same parameter controls as Qwen Generator

5. Kontext Generator (EigenAIKontextGeneratorNode)

Purpose: Image-to-image generation using Kontext model Input: Input image + Prompt + LoRA config + parameters Output: Generated image Usage:

• Upload or connect an input image
• Add text prompt for guidance
• Connect LoRA config for style enhancement
• Adjust generation parameters as needed

6. Upscaler (EigenAIUpscalerNode)

Purpose: Image upscaling and enhancement Input: Image + upscaling parameters Output: Upscaled image Usage:

• Connect input image from generator nodes
• Choose upscaling factor (2x or 4x)
• Apply ESRGAN upscaling for better quality

---

🔄 Basic Workflows

Standard Workflow

Text Node → LoRA Node → Generator Node → Upscaler → Output

Quick Workflow (Built-in LoRA)

Text Node → Generator Node → Output

Image-to-Image Workflow

Input Image → Kontext Generator → Upscaler → Output

---

---

🆘 Troubleshooting

• Error: "This action is not allowed with this security level configuration." Solution: Edit:

  ComfyUI/user/default/ComfyUI-Manager/config.ini
  

  Change:

  security_level = normal
  

  to:

  security_level = weak
  

  Then restart ComfyUI.

---

📄 License

MIT License

---

✨ Enjoy using the Eigen AI FLUX API Plugin! 🎨

---