Run a full local AI environment with:

• Ollama for model runtime (e.g., LLaMA 3)
• OpenWebUI as a chat interface for Ollama (accessible via your WSL IP address (e.g., http://:3000))
• LangChain + Chroma as a basic RAG pipeline for question answering over local documents

git clone https://github.com/wimjongman/local-ai-rag-stack.git
cd local-ai-rag-stack

If you have not installed Ollama on your host system (for CLI usage), install it first:

curl -fsSL https://ollama.com/install.sh | sh

Then pull the model:

ollama pull llama3

To start the containers:

docker-compose up -d

If you're running this inside WSL and can't reach http://localhost:3000 from Windows:

1. Run this in WSL to find your IP:

   ip addr show eth0 | grep inet

2. Use the resulting IP (e.g. http://172.20.5.234:3000) in your Windows browser.

Note: Ollama (port 11434) is usually directly accessible via localhost, but OpenWebUI (port 3000) may require access via the WSL IP address.

cd rag
python3 -m venv .venv
source .venv/bin/activate
pip install --upgrade pip
pip install -r requirements.txt
python main.py

Ask questions about your own documents locally 🚀

To remove containers, volumes, cache and start over:

./reset.sh