Text to Image Generator

A modern web application that generates images from text descriptions using Stable Diffusion AI model. Built with Streamlit and designed with a sleek dark theme interface.

Features

• Text-to-Image Generation: Convert text prompts into high-quality images using Stable Diffusion
• Modern UI: Dark theme with gradient styling and responsive design
• Real-time Processing: Instant image generation with loading indicators
• Public Access: Automatically exposed via ngrok for easy sharing
• GPU Support: Optimized for CUDA when available, falls back to CPU

Prerequisites

• Python 3.7+
• GPU with CUDA support (recommended) or CPU
• Internet connection for model downloads

Installation

Option 1: Google Colab (Recommended)

1. Upload all files to Google Colab
2. Run the Colab-specific runner:

!python run_colab.py

Option 2: Local Installation

1. Clone or download all files to your local machine
2. Install dependencies:

pip install -r requirements.txt

3. Run the application:

python main.py

Option 3: Quick Local Setup

Use the local runner that automatically checks and installs dependencies:

python local_run.py

Configuration

Environment Variables (Optional)

1. Copy .env.example to .env:

cp .env.example .env

2. Add your ngrok auth token (optional):

NGROK_AUTH_TOKEN=your_ngrok_auth_token_here

Note: Ngrok auth token is optional for basic usage but recommended for persistent URLs.

Usage

1. Start the Application:

   • Local: python main.py
   • Colab: python run_colab.py

2. Access the Interface: Click the ngrok URL that appears in the console

3. Generate Images:

   • Enter a descriptive text prompt
   • Click "Generate ➤"
   • Wait for the AI to process your request
   • View your generated image

Example Prompts

"A futuristic city skyline at night, cyberpunk style, neon lights"
"A serene mountain landscape with crystal clear lake, morning mist"
"Portrait of a wise old wizard, fantasy art style, detailed"
"Abstract art with vibrant colors and geometric patterns"
"A cute robot playing with butterflies in a magical garden"

Interface Features

• Responsive Design: Works on desktop and mobile devices
• Dark Theme: Easy on the eyes with blue gradient accents
• Input Validation: Prevents empty prompt submissions
• Loading States: Visual feedback during image generation
• Error Handling: User-friendly error messages
• Image Display: Rounded corners with glowing effects

🔧 Technical Details

Model Information

• Base Model: CompVis/stable-diffusion-v1-4
• Framework: Hugging Face Diffusers
• Precision: Float32 (configurable)

Performance Optimization

• Model Caching: Uses Streamlit's @st.cache_resource for efficient loading
• GPU Acceleration: Automatically detects and uses CUDA when available
• Background Processing: Runs Streamlit in a separate thread

Project Structure

text-to-image-generator/
├── main.py              # Main application entry point
├── app.py               # Auto-generated Streamlit app
├── requirements.txt     # Python dependencies
├── setup.py            # Package setup configuration
├── .env.example        # Environment variables template
├── .env                # Your environment variables (create from example)
├── local_run.py        # Local development runner
├── run_colab.py        # Google Colab runner
└── README.md           # This file

Dependencies

• streamlit>=1.28.0 - Web application framework
• diffusers>=0.21.0 - Stable Diffusion model implementation
• transformers>=4.34.0 - Hugging Face transformers library
• torch>=2.0.0 - PyTorch deep learning framework
• accelerate>=0.21.0 - Hugging Face acceleration library
• safetensors>=0.3.0 - Safe tensor serialization
• pyngrok>=6.0.0 - Python wrapper for ngrok tunneling
• python-dotenv>=1.0.0 - Environment variable management

Important Notes

First Run

• The first execution will download the Stable Diffusion model (~4GB)
• This may take several minutes depending on your internet connection
• Subsequent runs will be faster due to model caching

Hardware Requirements

• GPU: NVIDIA GPU with 4GB+ VRAM (recommended)
• CPU: Multi-core processor (fallback option)
• RAM: 8GB+ system memory
• Storage: 10GB+ free space for model files

Limitations

• Image generation can take 30-60 seconds per image
• CPU generation is significantly slower than GPU
• Some prompts may not produce expected results

Troubleshooting

Common Issues

"CUDA out of memory"

# Solution: The app will automatically fall back to CPU
# Or close other GPU-intensive applications

"Model download failed"

# Check internet connection
# Verify sufficient disk space (10GB+)
# Try running again (downloads can resume)

"Ngrok tunnel failed"

# Check if port 8501 is available
# Verify ngrok installation: pip install pyngrok
# Try restarting the application

"ModuleNotFoundError"

# Install missing dependencies
pip install -r requirements.txt

# Or use the local runner
python local_run.py

"Streamlit not found"

# Install Streamlit
pip install streamlit

# Or reinstall all dependencies
pip install -r requirements.txt

Getting Help

• Check the console output for detailed error messages
• Ensure all dependencies are properly installed
• Verify your Python version compatibility (3.7+)
• Make sure you have sufficient disk space for model downloads

Development

Running in Development Mode

# Install in development mode
pip install -e .

# Run with auto-reload
streamlit run app.py --server.runOnSave=true

Creating a Distribution

# Build the package
python setup.py sdist bdist_wheel

# Install locally
pip install dist/text-to-image-generator-1.0.0.tar.gz

Security Notes

• The application uses ngrok to create public tunnels
• Be cautious when sharing ngrok URLs publicly
• Consider using ngrok auth tokens for better security
• The app runs on your local machine and processes are isolated

License

This project uses the Stable Diffusion model which has specific licensing terms. Please review the model's license before commercial use.

Contributing

Feel free to submit issues, feature requests, or pull requests to improve this application.

Development Setup

1. Fork the repository
2. Create a virtual environment: python -m venv venv
3. Activate: source venv/bin/activate (Linux/Mac) or venv\Scripts\activate (Windows)
4. Install dependencies: pip install -r requirements.txt
5. Make your changes
6. Test thoroughly
7. Submit a pull request

Future Enhancements

• Multiple model support (SD 2.0, DALL-E, etc.)
• Batch image generation
• Image editing capabilities
• Custom model fine-tuning
• Advanced prompt engineering tools
• Image upscaling options
• Style transfer features
• API endpoint creation
• Docker containerization
• Database integration for image storage

📊 Performance Benchmarks

Hardware	Generation Time	Memory Usage
RTX 3090	15-30 seconds	8GB VRAM
RTX 3060	30-45 seconds	6GB VRAM
CPU (16 cores)	2-5 minutes	8GB RAM

Acknowledgments

• Hugging Face for the Diffusers library
• Stability AI for the Stable Diffusion model
• Streamlit for the web framework
• ngrok for public URL tunneling

Support

If you encounter any issues or have questions:

1. Check the troubleshooting section above
2. Look for similar issues in the project issues
3. Create a new issue with detailed information
4. Include your system specs and error messages

---

**Happy generating! **

Made with ❤️ by Zubair Hussain