SmartTextAnalysis-API

A Node.js REST API that uses AI to rewrite (rephrase/paraphrase) text input into natural, human-like language.
It leverages the LLaMA 3 70B Instruct model hosted by Groq via OpenRouter — one of the fastest available large language model (LLM) APIs.

Live Demo

Frontend: https://endearing-queijadas-237724.netlify.app/

Backend: https://smarttextanalysis-api.onrender.com

✨ Features

• 🔁 Rephrase any text into fluent, human-like English
• 🧠 Powered by meta-llama/llama-3-70b-instruct via OpenRouter API
• 💨 Fast inference using Groq’s ultra-low latency backend
• 📦 Modular structure with rephraser service
• ☁️ Supports future deployment and frontend integration
• 📜 Clean API response with original and rewritten text

🛠 Tech Stack

Part	Tech
Backend	Node.js, Express, Axios, dotenv
AI Model	LLaMA 3 (via OpenRouter & Groq)
Optional	MongoDB (for future history saving)
Frontend	Planned: React or Vue (optional)

🔌 API Endpoint

POST /api/rephrase

Request:

{ "text": "I live in a house near the mountains." }

Response:

{ "original": "I live in a house near the mountains.", "rewritten": "My home is nestled in the mountains." }

Getting Started

1. Clone the repository

git clone https://github.com/krystal-cpp/SmartTextAnalysis-API.git cd SmartTextAnalysis-API

2. Install dependencies

npm install

3. Add your .env file:

API_KEY_OPENROUTER=sk-or-xxxxxxxxxxxxxxxxxxxxxxxx

4. Run the server

node server.js The API will be available at: http://localhost:5000/api/rephrase

Roadmap

• Add style options (formal, casual, etc.)

• Store rephrase history in MongoDB

• React-based frontend interface

• Rate limiting / abuse protection

• User authentication (optional)

• Dockerization + CI/CD pipeline

Why This Project?

This project demonstrates how to:

• Integrate cutting-edge AI models (LLaMA 3 via OpenRouter)

• Build clean REST APIs in JavaScript

• Work with third-party LLM APIs

• Prepare for real-world backend developer roles