Vector Search API

A FastAPI application implementing vector similarity search.

Overview

This project demonstrates how to build a vector similarity search system using:

• PostgreSQL with pgvector extension for vector operations
• OpenAI/Ollama Embedding models for text vectorization
• FastAPI for the REST API interface
• Docker for containerization

Quick Start

1. Create .env in the root directory

   OPENAI_API_KEY=your_openai_api_key

2. Start the services

   docker compose up --build

API Endpoints

Visit http://localhost:8000/docs for the Swagger UI.

Development

• Update database schema: Edit postgres/schema.sql
• Add data: Use scripts/load_data.py
• Modify API: Edit files in api/