Run OpenMake in your own environment

Run this in a terminal. The default path prepares the host and installs the full OpenMake stack; it is not the app-only --minimal mode.

curl -fsSL https://raw.githubusercontent.com/openmake/openmake_llm/main/install.sh | bash

The macOS and Linux installers gather their questions first, then set up the required tools and the stack. The default source location is ~/.openmake/online/llm.

• Choose a DGX/vLLM host or an external provider such as OpenRouter. A provider needs its model ID and API key.
• Choose an access hostname or IP and whether to use internal HTTPS. The setup may request administrator privileges to install system tools.
• The full path sets up the app, PostgreSQL, Redis, a LiteLLM gateway, SearXNG, sandbox images, backups, and startup after reboot. Optional components depend on your answers.

When installation finishes, open the URL it prints and sign in with the administrator email and generated password. Change that password, then send one message and confirm that your chosen model answers.

• Check that the configured model appears in the composer and returns a response.
• The image below is a capture of an installed app. The hosted live demo runs separately and does not show your own deployment.

Close

For the default online environment, use the omk command installed under ~/.openmake/bin. If you set OMK_ROOT, use its bin/omk instead. The update command fetches code, builds, migrates, and restarts the environment.

~/.openmake/bin/omk env status online
~/.openmake/bin/omk env logs online
~/.openmake/bin/omk env update online

The full installer and app-only --minimal mode serve different needs. The repository README and installer --help document the current flags; use them before customizing a deployment. If a step fails or this guide drifts, report it in the open project.