Log in

Retrieve your license key and downloads. First time here? The password you enter becomes your account password.

Don't have an account? Sign up → Manage account / cancel subscription → Forgot your password? Reset it →
How setup works — from download to running (read me first)
  1. Install Docker. Windows & macOS: Docker Desktop (Windows users: Docker will set up WSL2 if prompted). Linux: Docker Engine — curl -fsSL https://get.docker.com | sh, then sudo usermod -aG docker $USER and log out and back in. Twhyne runs inside Docker so it works identically on every machine.
  2. Log in above to get your license key, then download the launcher for your system (buttons appear after login).
  3. Run the launcher — Windows: double-click twhyne-ai-setup.bat. macOS: open the DMG and double-click the app. Linux: from a terminal, chmod +x twhyne-ai-linux.sh && ./twhyne-ai-linux.sh (it needs a terminal to ask for your key). Paste your license key when asked. The first run downloads the AI models (about 13 GB, one time only) and the Twhyne app itself. This can take a while on slow connections — let it finish.
  4. Use the app. Your browser opens http://localhost:3000 automatically. The first answer can take a minute while a model loads into memory; after that it's faster.
  5. Updates are automatic: just run the launcher again. It checks for the newest version of Twhyne every time it starts and updates itself. Your models and knowledge bases are kept — no re-download.
  6. Your data stays on your machine. Models live in ~/.twhyne/models and your uploaded knowledge bases in ~/.twhyne/rag. Nothing you type or upload ever leaves your computer — the only internet call Twhyne makes is validating your license key.
  7. To stop: press Ctrl+C in the launcher window, or run docker stop twhyne-ai.
Requirements: 64-bit PC or Mac, 16 GB RAM recommended (12 GB available to Docker), and at least 25 GB free disk space. Trouble? Make sure Docker Desktop is running before launching, and that you have enough free disk — a full disk is the most common cause of startup failures.