🚀 Cascading Free AI API Flow

A robust Python system that automatically switches between multiple free AI API providers for reliable chat completions and text generation. Never worry about rate limits or downtime again!

✨ Features

• 🔄 Automatic Fallback: Seamlessly switches providers when one fails or hits limits
• 📊 Usage Tracking: Monitors daily usage across all providers
• ⚡ Exponential Backoff: Smart retry logic to handle temporary failures
• 🔧 Modular Design: Clean, maintainable code split across multiple files
• 📈 Real-time Stats: Monitor API consumption and performance
• 🛡️ Robust Error Handling: Comprehensive logging and graceful failures

🎯 Supported Free Providers (2025)

Provider	Daily Limit	Speed	Models
Groq	1,000 requests	⚡ Fastest	Llama 3.3 70B
Cerebras	1M tokens	🚀 High throughput	Llama 3.3 70B
OpenRouter	200 requests	🔀 Multi-model	Various free
Together AI	500 requests	🤝 Open source	Llama variants
Mistral	200 requests	🇫🇷 European	Mistral 7B
HuggingFace	1,000 requests	🤗 Community	Llama 3.3 70B
Fireworks	500 requests	🎆 Fast inference	Llama variants

🚀 Quick Start

1. Clone the Repository

git clone https://github.com/pratikm778/Free-API-setup.git
cd Free-API-setup

2. Install Dependencies

pip install -r requirements.txt

3. Set Up API Keys

cp .env.example .env
# Edit .env and add your API keys

4. Test Your Setup

python test_providers.py

4. Run Examples

python example.py

📁 Project Structure

cascading-ai-flow/
├── cascade.py          # Main cascading API client
├── providers.py        # Provider configurations
├── usage_tracker.py    # Usage tracking functionality
├── utils.py           # Utility functions
├── config.py          # Settings and configuration
├── example.py         # Usage examples
├── test_providers.py  # Provider testing script
├── requirements.txt   # Dependencies
├── .env.example      # Environment variables template
└── README.md         # This file

🔧 Basic Usage

from cascade import CascadingAPIClient

# Initialize client (uses all available providers)
client = CascadingAPIClient()

# Send a message
messages = [
    {"role": "user", "content": "What is machine learning?"}
]

response = client.chat_completion(messages)
print(response)

🔑 API Keys Setup

Get free API keys from these providers:

Provider	Signup Link	Notes
Groq	console.groq.com	Fastest inference
Cerebras	inference.cerebras.ai	High throughput
OpenRouter	openrouter.ai	Multiple models
Together AI	api.together.ai	Open source focus
Mistral	console.mistral.ai	European provider
HuggingFace	huggingface.co	Community models
Fireworks	fireworks.ai	Fast inference

💡 Tip: You don't need all keys! Start with 2-3 providers and add more later.

🧪 Testing

# Test all providers
python test_providers.py

# Quick test (individual providers only)
python test_providers.py --quick

📊 Advanced Usage

from cascade import CascadingAPIClient

client = CascadingAPIClient()

# Custom parameters
response = client.chat_completion(
    messages=messages,
    max_tokens=500,
    temperature=0.8,
    max_retries=3
)

# Get usage statistics
stats = client.get_usage_stats()
print(stats)

🔧 Configuration

Edit config.py to customize:

• Logging levels and file locations
• Default API parameters
• Retry and backoff settings
• Usage tracking options

📈 Monitoring

The system automatically:

• Tracks daily usage in usage_tracking.json
• Logs all activity to cascade_api.log
• Resets usage counters daily
• Provides real-time statistics

🐛 Troubleshooting

No providers available

• Check API keys in .env file
• Run python test_providers.py to verify setup

Rate limit errors

• System handles automatically with backoff
• Check usage with client.get_usage_stats()

API errors

• Check cascade_api.log for detailed errors
• Verify model names are correct

🤝 Contributing

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Add tests if applicable
5. Submit a pull request

📄 License

MIT License - use freely in your projects!

🙏 Acknowledgments

Thanks to all AI providers offering free tiers and the developer community for inspiration.

---

⭐ Star this repo if it helps you! ⭐