bigRAG
Getting Started

Installation

Install and run bigRAG using Docker Compose or from source.

Prerequisites

  • Docker and Docker Compose (recommended)
  • Python 3.12+ (for running from source)

Install

The fastest way to get started. One command starts the full stack:

docker compose up -d

This starts:

  • bigRAG API on port 6100 (Swagger docs at /docs)
  • Milvus vector database on port 19530
  • PostgreSQL for metadata on port 5433
  • Redis for the ingestion queue on port 6380

Pre-built images are published to Docker Hub:

docker pull yoginth/bigrag:latest

Start infrastructure services

docker compose up postgres redis milvus -d

Install and run the backend

cd api
python -m venv .venv && source .venv/bin/activate
pip install -e .
python -m bigrag.main \
  --database-url "postgres://bigrag:bigrag@localhost:5433/bigrag" \
  --milvus-uri "http://localhost:19530"

Development Mode

The easiest way to start everything for development:

./dev.sh

This script:

  1. Kills stale processes on port 6100
  2. Validates required commands (docker, python3, curl)
  3. Starts Docker services (Postgres, Redis, Milvus) and waits for readiness
  4. Creates a Python virtualenv and installs dependencies
  5. Starts the backend with auto-reload
  6. Gracefully stops everything on Ctrl+C

Verify

curl http://localhost:6100/health
# → {"status": "ok", "version": "0.x.x"}

Interactive Swagger docs are available at http://localhost:6100/docs once the server is running.

On this page