← Back to Help

Access Your Local LLM Server From Anywhere

Learn how to securely access your home LLM server (LM Studio or Ollama) from any device, anywhere.

What is Tailscale?

Tailscale is a zero-config VPN that creates a secure network between your devices. It allows you to access your home computer's services (like LM Studio or Ollama) from anywhere, as if you were on your home network.

Why Use Tailscale?

  • Access your local AI models from anywhere
  • Secure, encrypted connections
  • No complex network configuration needed

Step-by-Step Setup Guide

1. Set Up Tailscale on Your Computer

  1. Visit tailscale.com/download
  2. Download and install Tailscale for your operating system
  3. Sign in with your preferred account (Google, Microsoft, etc.)
  4. Allow Tailscale to configure your network settings

2. Install Tailscale on Your Mobile Device

  1. Install Tailscale from your device's app store
  2. Sign in with the same account used on your computer
  3. Accept the VPN configuration prompt

3. Find Your Computer's Tailscale IP

  1. On your computer, click the Tailscale icon
  2. Look for "Tailscale IP" (usually starts with 100.x.x.x)
  3. Note this IP address - you'll need it for 3sparks Chat

4. Configure 3sparks Chat

For LM Studio:

  1. Replace "localhost" with your Tailscale IP
  2. Keep the port number (usually 1234)
  3. Example: http://100.x.x.x:1234

For Ollama:

  1. Replace "localhost" with your Tailscale IP
  2. Keep the port number (11434)
  3. Example: http://100.x.x.x:11434

Important Notes

  • Your computer must be turned on and running LM Studio/Ollama
  • Both devices must be connected to the internet
  • Tailscale's free tier is sufficient for this use case
  • Response times may vary based on internet connection speed

Troubleshooting

Can't Connect to Server
  • Verify both devices show as connected in Tailscale
  • Confirm LM Studio/Ollama is running on your computer
  • Check if you can ping the Tailscale IP from your mobile device
Slow Response Times
Consider using a smaller model or check your internet connection speed on both devices.