Gettingstarted.

1. Open the Download page.
2. Pick your OS (Windows, macOS, or Linux).
3. Run the installer and follow the on-screen instructions.
4. Launch Typilot after installation.

The Ollama engine that runs the AI models ships inside the installer. There is nothing else to install.

1. On first launch, the onboarding wizard asks you to pick a model.
2. Choose llama3.2:3b- it's marked Recommended.
3. Wait for the download (a few minutes).

You can add or remove models later from the Models page. See 10 · Choosing AI models for picks beyond the default.

• Settings → Default model → pick the model you just pulled.
• Activation shortcut defaults to Ctrl + Space on Windows/Linux and Cmd + Space on macOS.
• Send key defaults to . (period). Press it after typing a command to submit.

See 07 · Keyboard shortcuts to rebind.

1. Open any app where you type - editor, browser, chat.
2. Press your activation shortcut.
3. Type: gen: write a hello world program in Python.
4. Press . to send.
5. Watch AI stream the response into the field.

• Activation feels weird at first. The shortcut is global - it works in every app. Try it in Notepad, Slack, your terminal.
• Privacy is structural.All inference is local. Your keystrokes stay on your machine even when you're offline.
• Start small. The 3b model handles ~80% of daily tasks. Add specialized models only when you hit their use cases.