# Ollama + Open WebUI

> Run open AI models on your own server with Ollama, and chat with them in Open WebUI.

Ollama runs open-weight language models behind a local API on 127.0.0.1:11434. We install it with Ollama's script as a systemd service, pull llama3.2:3b (about 2 GB) to start with, and run Open WebUI's image on 127.0.0.1:8080 with sign-ups off and an admin account whose password is in /root/open-webui-admin.txt.

## Details

- Installs on: Ubuntu and Debian; AlmaLinux, Rocky Linux and Fedora
- Smallest plan: Standard (2 vCPU, 4 GB RAM)
- Public ports: 80/tcp, 443/tcp
- Project site: https://ollama.com/

## After the server is ready

1. From your computer, run `ssh -L 8080:127.0.0.1:8080 root@<your IPv4>`, open `http://localhost:8080` and log in with the email and password from `cat /root/open-webui-admin.txt`.
2. Change the admin's email and password in Open WebUI's account settings.
3. Chat with llama3.2:3b, or download another model that fits your plan with `ollama pull <model>`.
4. For HTTPS on a domain, allow ports 80 and 443 in the panel firewall and run `vps9-https ai.example.com`.

## Questions

### How fast is Ollama without a GPU?

Slow compared with hosted AI. We haven't benchmarked it, and Ollama publishes no CPU figures. Small models (1B to 3B) are usable for short questions on Standard; 7B and 8B models fit in Extreme's memory but answer noticeably slower. Longer conversations use more memory and time.

### Which models fit my plan?

Count the model's download size plus room for its context and Open WebUI. Standard (4 GB): llama3.2:1b (1.3 GB) or llama3.2:3b (2 GB). Extreme (8 GB): those, or 7B and 8B models such as llama3.1:8b (4.9 GB). Ollama's library lists every model's size.

### Can my own apps use the models?

Yes, through Ollama's API, but it has no authentication: reach it through an SSH tunnel (`ssh -L 11434:127.0.0.1:11434 root@<your IPv4>`) rather than opening the port. Open WebUI also offers an API with keys you create in its settings.

### Do my prompts leave the server?

Not for models running in Ollama. Open WebUI can also connect to hosted providers if you add their API keys; prompts you send to those models go to that provider.

## Order it

Pick Ollama + Open WebUI in the order form, or use `"app": "ollama"` in the API or `vps9 servers create --app ollama`. See [all one-click apps](https://vpsnine.com/apps) and [plans](https://vpsnine.com/plans).