🧠 OpenRouter AI Examples (TypeScript)

This repository showcases how to use OpenRouter for interacting with LLMs in a Node.js (TypeScript) environment using streaming, prompt templates, and document-based Q&A.

---

📦 Features

• 🤖 Call OpenRouter AI models with streaming support.
• 📄 Ask questions from a PDF document.
• 🛠️ Fully typed with TypeScript.
• 📂 Document-based LLM retrieval with npx CLI support.

---

🚀 Getting Started

1. Clone the repo

git clone https://github.com/sfvishalgupta/OpenRouterAIExample.git
cd OpenRouterAIExample

2. Install dependencies

npm install

3. Set up environment variables

Create a .env file in the src folder:

OPEN_ROUTER_API_KEY=your_openrouter_api_key

Example 1: Ask a Question

Run the below Command

npx ts-node src/askQuestion.ts <question>

Example 2: Ask a Question from a PDF

Place your pdf inside src/documents folder

Run the below Command

npx ts-node src/askQuestionFromPDF.ts <pdfPath> <question>

Example 3: Add PDF to Qdrant DB

Place your pdf inside src/documents folder

Run the below Command

# Run below command to start Docker.
docker run -p 6333:6333 -p 6334:6334 qdrant/qdrant

npx ts-node src/syncDataToVectorDB.ts <pdfPath> <index_name>

Example 2: Ask a Question from a Qdrant DB

Run the below Command

npx ts-node src/askQuestionFromVector.ts <index_name> <question>

🧠 Tech Stack • Node.js • TypeScript • OpenRouter • PDF-parse for reading PDFs

📬 License MIT — feel free to use and adapt!