This project turns an old laptop into a fully functional AI server that can serve LLMs locally and access them globally using Cloudflare Tunnel on API . A simple Streamlit app provides a friendly frontend to interact with the model via API .

🔗 Blog post: Run a Local AI Server for Free with Ollama & Cloudflare Tunnel

• Personal chatbot
• Code generation tool
• Lightweight AI backend for your apps
• Home AI server on a budget

• A Old laptop with Linux (preferably with 4GB+ RAM)
• Python 3.9+
• Ollama
• Cloudflared

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

$ ollama pull tinyllama  # or any other model like deepseek-llm

$ OLLAMA_HOST=0.0.0.0 ollama serve

# Or run on a custom port

$ OLLAMA_HOST=0.0.0.0 OLLAMA_PORT=11435 ollama serve

$ wget https://github.com/cloudflare/cloudflared/releases/latest/download/cloudflared-linux-amd64 -O cloudflared

$ chmod +x cloudflared

$ cloudflared --version

$ cloudflared tunnel --url http://localhost:11435

This will give you a public URL like:

https://your-unique-subdomain.trycloudflare.com

Install dependencies:

pip install -r requirements.txt

Run app:

streamlit run app.py

Make sure to update your app.py with the correct Cloudflare URL.

curl https://your-unique-subdomain.trycloudflare.com/api/generate -d '{
  "model": "deepseek-llm:1.5b",
  "prompt": "Write a Python function to reverse a string"
}'

• Use htop or glances to monitor usage
• Check disk with lsblk and df -h
• Stick to lightweight models for faster inference

local-ai-server/
├── app.py                # Streamlit frontend
├── requirements.txt      # Python deps
├── README.md             # You are here

Built with ❤️ by Raghul M

Inspired by a simple idea: "What if I could run AI server without the cloud?"