It just runs
Once installed, Ollama keeps a small service running so apps can talk to your model. You can also use it from other programs without reconfiguring anything.
Getting started with Ollama
Ollama is the easiest way to run a local model. You install it, type one command to download a model, and then chat with it in your terminal.
About 1 min of reading
Data Flow
Ollama is installed, a model is pulled with one command, and then you chat with it directly in the terminal.
1
Install Ollama
↓
2
Pull a model
↓
3
Run it
↓
4
Chat in terminal
Why this works
When you have a choice
Pick a model
Ollama has many models to choose from. Start with a small one like llama3.2 if your computer is not powerful. You can always try bigger ones later.
Examples
Three commands
1. Install Ollama from the website
2. ollama run llama3.2
3. Start typing your questionThat is all it takes to have a working local model. Ollama handles the hard parts for you.
Pull more models
ollama pull mistral
ollama pull phi3
ollama listYou can download as many models as you want. Each lives separately, so you can switch between them with ollama run <name>. Keep small ones for quick tasks and pull bigger models when you need depth.
Keeps running in the background
Ollama stays active in the background after you install it. Other programs can send requests to it without you opening a terminal.
This is why it is a popular choice: one install, many apps. Your editor, a notes app, or a custom script can all talk to the same model.
Words this lesson uses
Each one is defined here on this site. Open the entry for the full definition and the other words it points at.
← What is a local LLM?
Previous Lesson
What is a local LLM?