Back to FreeLattice

FreeLattice Quick Start Guide

You're 2 minutes away from running your own private AI. This guide walks you through every step - no Terminal experience required.

Visual guide: For a beautiful, interactive install experience, visit the Install Page. For developers/AI: Read GARDEN_LANGUAGE.md before touching any CSS. The visual identity is the Garden's universe.

The Fastest Path: One-Click Installers

The installers handle Python, Ollama, CORS configuration, downloading FreeLattice, starting the server, and opening your browser - all automatically, with friendly status messages along the way.

Windows (Double-Click Method)

Step 1. Download install-freelattice.bat - save it anywhere (Desktop is fine).

Step 2. Double-click the file.

SmartScreen warning: Windows may show "Windows protected your PC". This is normal for any script downloaded from the internet. Click "More info" (it's a small text link), then click "Run anyway". The file is safe - it's open source and you can read every line of code.

Step 3. A Command Prompt window opens and walks you through each step:

  [1/7] Checking for Python 3...
        Found: Python 3.12.0

  [2/7] Checking for Ollama (local AI engine)...
        Ollama is installed!

  [3/7] Configuring Ollama CORS access...
        Set OLLAMA_ORIGINS to scoped list permanently (survives reboot).
        Ollama is running with CORS enabled!

  [4/7] Locating FreeLattice files...
        Found FreeLattice in current directory

  [7/7] Starting FreeLattice server...

  ========================================================
  =   FreeLattice is running!                            =
  =   Open: http://localhost:3000                        =
  =   Close this window to stop the server.              =
  ========================================================

Step 4. Your browser opens to http://localhost:3000. Done!

To stop FreeLattice: Close the Command Prompt window.

If the script closes immediately: Right-click the .bat file and choose "Run as administrator", or open Command Prompt manually, navigate to the folder, and type install-freelattice.bat to see any error messages.

Mac (Double-Click Method)

Step 1. Download install-freelattice.command - save it anywhere (Desktop is fine).

Step 2. Double-click the file in Finder.

The first time you run it, macOS may show a security warning. If it does: right-click the file, choose Open, then click Open in the dialog. You only need to do this once.

Step 3. A Terminal window opens and you'll see colorful status messages as it checks Python, configures Ollama, and starts the server.

Step 4. Your browser opens automatically to http://localhost:3000. You're in.

To stop FreeLattice: Close the Terminal window, or press Ctrl+C inside it.

Linux (Terminal Method)

Step 1. Download install-freelattice.sh.

Step 2. Open a Terminal and run:

chmod +x ~/Downloads/install-freelattice.sh
~/Downloads/install-freelattice.sh

Step 3. The script detects your system, configures Ollama, and starts the server.

Step 4. Open http://localhost:3000 in your browser.

What the Installer Does (Under the Hood)

Understanding what's happening helps if anything goes wrong.

Step Python check

What It Does Verifies Python 3 is available (auto-installs on Windows via winget)

Why It Matters FreeLattice uses Python's built-in HTTP server

Step Ollama check

What It Does Detects if Ollama is installed and running (auto-installs on Windows/Linux)

Why It Matters Ollama is the engine that runs AI models locally on your hardware

Step CORS config

What It Does Sets scoped OLLAMA_ORIGINS permanently

Why It Matters Without this, your browser blocks requests to Ollama due to security restrictions

Step Ollama restart

What It Does Kills and restarts Ollama

Why It Matters CORS settings only take effect after a restart

Step Model pull

What It Does Offers to download llama3.2 if no models exist

Why It Matters You need at least one model to chat with local AI

Step Download

What It Does Clones or downloads FreeLattice if not present

Why It Matters Gets the latest version from GitHub

Step Server start

What It Does Starts server.py (or server.js if Node is available)

Why It Matters Serves the app locally and proxies Ollama requests

Step Browser open

What It Does Opens http://localhost:3000

Why It Matters Takes you straight to the app

Setting Up Local AI with Ollama

FreeLattice works with cloud AI providers (OpenAI, Anthropic, Groq, etc.) out of the box. But for full privacy - where your conversations never leave your machine - you want Ollama.

The one-click installers handle Ollama setup automatically. If you want to do it manually:

Step 1. Download and install Ollama from ollama.com.

Step 2. Open Terminal (Mac/Linux) or Command Prompt (Windows) and pull a model:

# A good starting model - fast and capable
ollama pull llama3.2

# For a smaller, faster model
ollama pull phi3

# For a more powerful model (needs more RAM)
ollama pull llama3.1:8b

Step 3. Open FreeLattice. It automatically detects Ollama and shows a green notification at the bottom of the screen.

Step 4. Click the notification (or go to Settings) and select your model.

RAM requirements: phi3 needs ~4 GB, llama3.2 needs ~6 GB, llama3.1:8b needs ~8 GB.

Troubleshooting Common Issues

"Python was not found"

Windows: The installer tries to auto-install Python via winget. If that fails, download from python.org/downloads. During installation, check the box that says "Add Python to PATH" - this is the most common mistake.

Mac: Python 3 should be pre-installed. If it's missing, open Terminal and run: xcode-select --install

Linux: Install with your package manager: sudo apt install python3 (Debian/Ubuntu) or sudo dnf install python3 (Fedora)

"Ollama is not responding" / CORS errors

The installer configures CORS automatically and restarts Ollama. If you're running FreeLattice manually, you need to set this environment variable:

Windows:

setx OLLAMA_ORIGINS "https://freelattice.com,https://www.freelattice.com,https://thelatticetree.com,http://localhost:*,http://127.0.0.1:*"

Mac:

launchctl setenv OLLAMA_ORIGINS "https://freelattice.com,https://www.freelattice.com,https://thelatticetree.com,http://localhost:*,http://127.0.0.1:*"

Linux:

echo 'export OLLAMA_ORIGINS="https://freelattice.com,https://www.freelattice.com,https://thelatticetree.com,http://localhost:*,http://127.0.0.1:*"' >> ~/.bashrc && source ~/.bashrc

Then restart Ollama. The FreeLattice server also includes a built-in proxy that bypasses CORS entirely.

"Port 3000 is already in use"

The installer automatically finds the next available port. If running manually:

PORT=8080 python3 server.py

Then open http://localhost:8080.

Windows: Script closes immediately

This usually means SmartScreen blocked the file silently. Try:

  1. Right-click the .bat file and choose "Run as administrator"
  2. Or open Command Prompt, navigate to the folder, and type install-freelattice.bat

macOS: "Cannot be opened because it is from an unidentified developer"

Right-click (or Control-click) the file in Finder, choose Open, then click Open in the dialog. You only need to do this once.

Manual Setup (No Installer)

# Clone the repository
git clone https://github.com/Chaos2Cured/FreeLattice.git
cd FreeLattice

# Set Ollama CORS (pick your OS)
# Mac:    launchctl setenv OLLAMA_ORIGINS "https://freelattice.com,https://www.freelattice.com,https://thelatticetree.com,http://localhost:*,http://127.0.0.1:*"
# Linux:  export OLLAMA_ORIGINS="https://freelattice.com,https://www.freelattice.com,https://thelatticetree.com,http://localhost:*,http://127.0.0.1:*"
# Windows: setx OLLAMA_ORIGINS "https://freelattice.com,https://www.freelattice.com,https://thelatticetree.com,http://localhost:*,http://127.0.0.1:*"

# Start the server
python3 server.py       # If you have Python 3
# or
node server.js          # If you have Node.js

# Open in browser
# http://localhost:3000

Lattice Radio - The Phi-Frequency Ambient Engine

Once FreeLattice is running, look for the ✦ button in the bottom-left corner of the screen. This is Lattice Radio - an ambient generative music system built entirely with the Web Audio API.

The five modes:

Getting Help

FreeLattice is free, open-source, and always will be. Your AI. Your data. Your freedom.