Skip to content

README.md

Latest commit

 

History

History
52 lines (40 loc) · 1.09 KB

README.md

File metadata and controls

52 lines (40 loc) · 1.09 KB

InputX

InputX is a high-speed API gateway for AI inference, enabling seamless, low-latency access to large language models through a simple and scalable interface.

🚀 Features

  • 🔐 API key-based authentication
  • ⚡ Fast and lightweight Node.js backend using Fastify
  • 🧠 OpenAI integration (with more models coming soon)
  • 📊 Usage logging for metering and billing
  • 🧱 Ready for deployment on Vercel, AWS Lambda, or Fly.io

📦 Getting Started

git clone https://github.com/yourusername/inputx.git
cd inputx
npm install
cp .env.example .env
# Add your OpenAI API key to .env
npm start

🔑 API Usage

Endpoint

POST /inference

Headers

Authorization: Bearer <your-api-key>

Body

{
  "prompt": "What is the capital of France?",
  "model": "gpt-3.5-turbo"
}

🧱 Roadmap

  • Add support for Ollama / local models
  • Token-based usage tracking and billing
  • Web3 wallet-based access and payments
  • Caching layer for repeated queries

📄 License

MIT - see LICENSE

InputX: Serve AI like infrastructure.