Video course · 14 chapters · 30 min · certificate
Build a Private Local AI Server with Ubuntu, Docker & Ollama
Build a private AI server at home: why local AI matters, hardware and CUDA GPUs, installing Ubuntu, updates, Docker and Ollama, Open WebUI, an optional Wi-Fi hotspot, starting on boot and fixing errors, installing models and setting up multiple users.
What you'll learn
- Explain the benefits of local AI
- Choose suitable hardware
- Install Ubuntu, Docker and Ollama
- Run Open WebUI as a chat interface
- Auto-start the server and fix common errors
- Install models and add users
Chapters
14 chapters · 30:12-
2:03
01Intro Members
Intro and USB prep
Running your own AI server, and preparing tools.
-
2:10
02Why Members
Why local AI matters
Privacy versus online AI services.
-
2:43
03Hardware Members
Hardware overview
GPUs with CUDA cores.
-
1:42
04OS Members
Installing Ubuntu
Booting from USB and installing.
-
1:57
05Tips Members
Research and struggles
Lessons from failed installs.
-
2:35
06Setup Members
Server updates
Updating the server; password tips.
-
2:02
07Install Members
Docker and Ollama
Installing Docker and Ollama.
-
1:30
08Interface Members
Open WebUI
A ChatGPT-like interface for your models.
-
3:08
09Network Members
Optional Wi-Fi hotspot
Letting others connect via a hotspot.
-
2:13
10Ops Members
Start on boot and Error 500
Auto-start and fixing an NVIDIA error.
-
1:34
11Models Members
Installing models
Downloading models in Open WebUI.
-
2:02
12Models Members
Model setup
Configuring a model.
-
3:14
13Users Members
Multiple users
Letting several people use the AI.
-
1:19
14Wrap-up Members
Final thoughts
What's not covered and next steps.
Jobs this course suits
Our AI checked this course against 500 jobs; these get the most out of it. Each job links to its learning path.