⚡ API Latency Monitor

Real-time response time comparison for LLM API providers. Monitor OpenAI, Anthropic, Google AI, Azure, Cohere, and Hugging Face APIs with beautiful visualizations and detailed statistics.

🎯 Features

• Real-time Monitoring: Track API response times across multiple providers simultaneously
• Three Test Modes:
  • 🎯 Ping Test: Quick endpoint reachability check (no API key required)
  • 📝 Simple Request: Minimal API request simulation
  • 🔐 Full Request: Complete authenticated API calls (requires keys)
• Visual Charts: Line charts showing latency history over time
• Detailed Statistics: Min, max, average, success rates, and more
• Flexible Intervals: Test every 5 seconds to 1 minute, or manual testing
• Zero Dependencies: Pure JavaScript, HTML, and CSS - no frameworks
• Privacy First: All tests run in your browser, no data sent to external servers
• Open Source: MIT licensed, inspect and modify as needed

🚀 Quick Start

1. Clone or Download:

   git clone https://github.com/Turtles-AI-Lab/api-latency-monitor.git
   cd api-latency-monitor

2. Open in Browser:

   # Just open index.html in any modern browser
   open index.html  # macOS
   start index.html # Windows
   xdg-open index.html # Linux

3. Start Monitoring:

   • Select test mode (Ping recommended for quick checks)
   • Choose test interval
   • Click "Start Monitoring"
   • Watch the magic happen!

📊 Monitored Providers

• OpenAI - GPT models
• Anthropic - Claude models
• Google AI - Gemini models
• Azure OpenAI - Microsoft-hosted OpenAI
• Cohere - Command models
• Hugging Face - Inference API

🎮 Usage

Ping Test Mode (Recommended)

No API keys required. Simply checks if endpoints are reachable.

// Tests are conducted via browser fetch with no-cors mode
// Perfect for uptime monitoring

Simple Request Mode

Simulates basic API requests. Currently uses ping test as fallback.

Full Request Mode

Requires API keys for authenticated requests. Store keys locally in browser.

📈 Understanding the Data

• Current: Latest recorded latency
• Average: Mean latency across all successful tests
• Min/Max: Fastest and slowest response times
• Success Rate: Percentage of successful requests
• Fast: < 200ms
• Normal: 200-500ms
• Slow: > 500ms

🔒 Privacy & Security

✅ All tests run locally in your browser ✅ No data sent to external servers ✅ API keys stored in browser localStorage only ✅ No tracking, analytics, or cookies ✅ Open source - verify yourself

🛠️ Technical Details

Built With:

• Pure JavaScript (ES6+)
• HTML5 Canvas for charts
• CSS3 for styling
• No external dependencies

Browser Support:

• Chrome/Edge 90+
• Firefox 88+
• Safari 14+
• Any modern browser with ES6 support

📁 Project Structure

api-latency-monitor/
├── index.html          # Main HTML structure
├── css/
│   └── style.css       # All styling
├── js/
│   ├── providers.js    # API provider configurations
│   ├── monitor.js      # Core monitoring logic
│   ├── chart.js        # Chart visualization
│   └── app.js          # Main application logic
├── README.md
├── LICENSE
└── .gitignore

🤝 Contributing

Contributions welcome! Please feel free to submit a Pull Request. For major changes, please open an issue first.

Ideas for contributions:

• Add more API providers
• Implement authenticated request modes
• Add export functionality (CSV, JSON)
• Dark mode support
• Mobile app version
• Historical data persistence

📝 License

MIT License - see LICENSE file for details

🐢 About

Built by Turtles AI Lab - Free tools for AI developers

🔗 Related Tools

• LLM Cost Calculator - Compare AI API costs
• Prompt Template Library - Battle-tested AI prompts
• Email-to-Ticket Parser - Extract ticket info from emails

📧 Contact

Questions or feedback? Email: jgreenia@jandraisolutions.com

---

⭐ Star this repo if you find it useful!