Self-Host a local AI platform! Ollama + Open WebUI

Check out Twingate and supercharge your security: https://bit.ly/3Y1OaZi

In this video, I'll show you my new self-hosted AI platform deployed in my HomeLab using the free and open-source Ollama platform. I'll walk you through setting up OpenWebUI for an easy-to-use web interface with advanced features even ChatGPT might envy—and securing it with Traefik and Authentik. Plus, I’ll share some valuable tips to avoid common pitfalls when building your own local AI server.

References

- Ollama GPU Requirements: https://github.com/ollama/ollama/blob/main/docs/gpu.md

-AMD ROCm: https://rocm.docs.amd.com/projects/radeon/en/latest/docs/install/native_linux/install-radeon.html

Learn more

- Traefik: https://www.patreon.com/posts/114924172

- Authentik: https://www.patreon.com/posts/100779796

________________

💜 Support me and become a Fan!

→ https://christianlempa.de/patreon

→ https://www.youtube.com/channel/UCZNhwA1B5YqiY1nLzmM0ZRg/join

💬 Join our Community!

→ https://christianlempa.de/discord

👉 Follow me everywhere

→ https://christianlempa.de

________________

Read my Tech Documentation

https://christianlempa.de/docs

My Gear and Equipment

https://christianlempa.de/kit

________________

Timestamps:

00:00 Introduction

02:38 Hardware Requirements

06:58 Software Planning

08:35 Problems with Proxmox…

10:55 Installing a new LXC Container

14:44 Install Ollama on Linux

17:41 Ollama basics

19:56 Install OpenWeb UI

27:37 OpenWebUI basics

31:05 Using AI models

35:01 Web Searching

37:18 Why I still don’t trust AI

________________

Links can include affiliate links.