How to install Ollama + Open WebUI on a Cloud VPS

Ollama with Open WebUI runs large language models locally on your server with a polished chat interface, no API costs and full privacy.

Running it on your own server keeps your data and API keys with you, with no SaaS usage limits and predictable hourly billing.

One-click install on cloudlinux.ro

The fastest way: Ollama + Open WebUI installs automatically at the first boot of a new server.

1. Create a Cloud VPS from the control panel.
2. In the configuration step, pick Ollama + Open WebUI from the app catalog.
3. Start the server. The install runs in the background for a few minutes and its status shows on the server page.
4. Access details and generated passwords are saved on the server in /root/app-credentials.txt. The app will be available on port 3000.

Manual install on any Ubuntu server

If you prefer doing it yourself, below is the exact script our automatic install runs. Execute it as root on a fresh Ubuntu server (24.04 recommended). It stops at the first error, and when it finishes you will find the access details in /root/app-credentials.txt.

export DEBIAN_FRONTEND=noninteractive
curl -fsSL https://get.docker.com | sh
curl -fsSL https://ollama.com/install.sh | sh
systemctl enable --now ollama
docker run -d --name open-webui --restart=always -p 3000:8080 \
  --add-host=host.docker.internal:host-gateway \
  -v open-webui:/app/backend/data \
  -e OLLAMA_BASE_URL=http://host.docker.internal:11434 \
  ghcr.io/open-webui/open-webui:main
ollama pull llama3.2:3b || true
IP=$(hostname -I | awk '{print $1}')
printf 'Open WebUI: http://%s:3000\nCreeaza contul de administrator la prima accesare. Modelul llama3.2:3b este preinstalat.\n' "$IP" > /root/app-credentials.txt

You can also save this script as a recipe in your account and reuse it on any new server.