Skip to content

Install Terere Harness

Installing, updating and uninstalling Terere Harness on Windows and Mac, and where it keeps your data.

What you need

  • Windows 10 or Windows 11, 64-bit, or a Mac with Apple silicon (M1 or later) and macOS 13 or later. Intel Macs aren’t supported yet.
  • About 2 GB of free disk space.
  • An AI model to connect: a hosted API with a key, or a model running on your computer (for example with Ollama or LM Studio). See Connect a model.

You don’t need a Terere account.

Install on Windows

  1. Download the installer from the download page.
  2. Open the downloaded file. If Windows SmartScreen shows “Windows protected your PC”, select More info, then Run anyway. Preview builds aren’t code-signed yet, so the publisher shows as “Unknown publisher”. To confirm the file is ours, compare its checksum with the one on the download page first.
  3. Choose where to install it. The installer suggests a folder in your user profile, so you don’t need administrator rights.
  4. When it finishes, open Terere Harness from the Start menu.

Install on a Mac

  1. Download the disk image from the download page. It runs on Apple silicon: About This Mac in the Apple menu shows a Chip line, such as Apple M1. A Processor line means an Intel Mac, which isn’t supported yet.
  2. Open the disk image and drag Terere Harness into the Applications folder.
  3. Open Terere Harness from Applications. Preview builds aren’t notarised by Apple yet, so the first time macOS says it can’t check the app. Choose Done, open System Settings, then Privacy & Security, scroll down and choose Open Anyway, then confirm. On macOS 13 or 14 the first message looks different and offers OK instead of Done; the rest is the same. To confirm the file is ours first, compare its checksum with the one on the download page.
  4. If you’d rather use Terminal, this clears the download flag macOS checks: xattr -dr com.apple.quarantine "/Applications/Terere Harness.app"

First launch

The first window asks you to connect a model. Select Connect a model, enter your endpoint’s details and select Save and continue. Connect a model lists the values for common providers.

The Connect your AI endpoint form, filled in for Ollama: Base URL http://localhost:11434/v1, API type OpenAI-compatible (Chat Completions), API key left empty, Model ID qwen3:8b, with Save and continue and Set up later buttons.
The first-launch form, filled in for Ollama on your own computer.

Select Set up later to skip this step. You can add a model at any time in Settings, then Models.

The message box of a new session saying “Connect a model in Settings → Models to start”, with a Connect a model button where the model name usually is.
Until a model is connected, the message box points you to Settings, then Models.

Updates

Preview builds don’t update themselves yet. To update, download the new version from the download page: on Windows, run the new installer; on a Mac, drag the app into Applications over the old copy. Your settings and sessions stay.

Where your data lives

Everything stays on your computer:

  • Settings, sessions and saved API keys: %USERPROFILE%\.terere on Windows, ~/.terere on a Mac
  • App data, such as logs and window state: %APPDATA%\@terere\desktop on Windows; on a Mac, ~/Library/Application Support/@terere/desktop, with logs in ~/Library/Logs/@terere/desktop

Your prompts and code go from the app straight to the endpoint you connect. They never pass through Terere. If you connect the app to an Enterprise organisation, only token counts, and any facts you choose to share, encrypted, go to terere.ai.

Uninstall

  1. Open Settings, then Apps, then Installed apps.
  2. Find Terere Harness, open its menu and select Uninstall.
  3. To remove your settings, sessions and saved keys as well, delete the %USERPROFILE%\.terere folder.

On a Mac, drag Terere Harness from Applications to the Bin. To remove your settings, sessions and saved keys as well, delete the ~/.terere folder.