A modern AI image generation web application that supports multiple AI platforms, providing real-time image generation, download, and management features.

Deploy with EdgeOne Pages.

• 🎯 Multi-Platform Support - Integrated with mainstream AI platforms like Hugging Face, Nebius, Replicate
• 🚀 Real-time Generation - Support for real-time image generation with progress indicators
• 💾 One-Click Download - Support for image download with automatic filename generation
• 🎨 Multiple Models - Support for various AI models including SDXL, Flux, Pixel Art, and more
• 📱 Responsive Design - Perfect adaptation for desktop and mobile devices

• Next.js 15.3.4 - React full-stack framework
• React 19.0.0 - User interface library
• TypeScript 5.0 - Type-safe JavaScript
• Tailwind CSS 4.0 - Utility-first CSS framework
• TDesign React - Enterprise-class UI component library

• EdgeOne Functions - Edge computing functions
• Multi-Platform API Integration - Hugging Face, Nebius, Replicate

• Node.js 18.0 or higher
• npm or yarn package manager
• Valid AI platform API tokens

1. Clone the Repository

   git clone <repository-url>
   cd generate-graph
   

2. Install Dependencies

   npm install
   # or
   yarn install
   

3. Configure Environment Variables Create a .env file and add the following configuration (not all required):

   # Hugging Face API Token
   HF_TOKEN=your_huggingface_token_here
   
   # Nebius API Token  
   NEBIUS_TOKEN=your_nebius_token_here
   
   # Replicate API Token
   REPLICATE_TOKEN=your_replicate_token_here
   
   # OPENAI API Key
   OPENAI_API_KEY=your_openai_api_key_here   
   
   # FAL API Key
   FAL_KEY=your_fal_key_here
   

   The EdgeOne Pages deployment console environment variable configuration is consistent with the above .env file.

4. Start Development Server

   npm run dev
   # or
   yarn dev
   

5. Access the Application Open your browser and visit http://localhost:3000

1. Visit Hugging Face
2. Register and login to your account
3. Go to Settings > Access Tokens
4. Create a new access token
5. Copy the token to the HF_TOKEN environment variable

1. Visit Nebius Studio
2. Register and login to your account
3. Go to API settings page
4. Generate an API key
5. Copy the key to the NEBIUS_TOKEN environment variable

1. Visit Replicate
2. Register and login to your account
3. Go to Account Settings
4. Create an API token
5. Copy the token to the REPLICATE_TOKEN environment variable

1. Visit OpenAI
2. Register and login to your account
3. Go to API Keys
4. Create a new API key
5. Copy the key to the OPENAI_API_KEY environment variable

1. Visit FAL
2. Register and login to your account
3. Go to API Keys
4. Create an API key
5. Copy the key to the FAL_KEY environment variable

The application automatically checks the availability of tokens for each platform and displays the status in the interface:

• ✅ Available - Token is configured and valid
• ❌ Missing - Token is not configured or invalid

1. Select Model

   • Choose from available AI models in the left panel
   • Different models support different image styles and generation effects

2. Enter Prompt

   • Input image description in the text box
   • Supports both Chinese and English descriptions
   • The system automatically adds model style suffixes

3. Generate Image

   • Click the "Generate" button to start generation
   • Real-time display of generation progress and time
   • Automatically displays results when generation is complete

4. Download Image

   • Hover over the image to show download button
   • Click download button to save the image
   • Automatically generates filename containing the prompt

Endpoint: POST /v1/generate

Request Body:

{
  "image": "Image description text",
  "platform": "huggingface",
  "model": "Model identifier"
}

Response:

{
  "success": true,
  "prompt": "Original prompt",
  "imageData": "Base64 encoded image data or URL",
  "message": "Generation success message"
}

Endpoint: GET /v1/token-status

Response:

{
  "hfToken": true,
  "nebiusToken": true,
  "replicateToken": false
}

All API endpoints include unified error handling:

• 400 - Request parameter errors or content violations
• 429 - Request rate limit exceeded
• 500 - Internal server error

generate-graph/
├── functions/                 # EdgeOne Functions
│   ├── v1/
│   │   ├── generate/         # Image generation API
│   │   │   ├── index.js      # Main processing logic
│   │   │   ├── fetch_utils.js # API call utilities
│   │   │   └── nfsw_limit.js # Content filtering
│   │   └── token-status/     # Token status API
│   └── helloworld/           # Example function
├── src/
│   ├── components/           # React components
│   │   ├── ImageDisplay.tsx  # Image display component
│   │   └── ModelDropdown.tsx # Model selection component
│   ├── pages/               # Next.js pages
│   │   ├── _app.tsx         # Application entry
│   │   ├── _document.tsx    # Document configuration
│   │   └── index.tsx        # Main page
│   └── styles/              # Style files
│       └── globals.css      # Global styles
├── public/                  # Static assets
├── package.json             # Project configuration
├── next.config.ts           # Next.js configuration
├── tailwind.config.js       # Tailwind configuration
└── README.md               # Project documentation