Ollama Essentials — Getting Started

0Orientation & objectives

This module introduces Ollama from first contact: what it is, how it installs, the three interfaces for talking to it, the commands that manage models on disk and in memory, and how a Modelfile customizes a model. The aim is a working command of day-one operation, not an exhaustive reference.

Learning objectives. On completion a participant will be able to describe what Ollama is and install it, use the command line to run and manage models, choose between the CLI, the REST API, and the SDKs, and build a customized model from a Modelfile.

Format	Self-paced; approximately 20 minutes
Formative questions	8 multiple-choice, distributed (feedback, not scored)
Final assessment	8 multiple-choice questions; pass mark 75%
Sources	Official Ollama documentation (README, Quickstart, CLI Reference, Modelfile Reference), 2026-07-07

A companion module goes deeper on the context window and coding-tool integrations. This one stays at the general-introduction level.

1What Ollama is, and installing it

Learning objective. State what Ollama does and how it is installed, and place it within its ecosystem.

Ollama is local large-language-model infrastructure built on llama.cpp. Its own scope stays narrow — model serving — while a large third-party ecosystem builds interfaces, agent frameworks, and RAG stacks around it. Open WebUI is the largest single example of that pattern.

Installation is a one-line script on macOS and Linux, and an installer on Windows:

curl -fsSL https://ollama.com/install.sh | sh

After installing, ollama run gemma4 downloads the model if needed and starts an interactive chat; /bye exits. Running ollama on its own reaches an interactive menu — chat with a model, or launch an integrated tool.

The :cloud tag. For models too large to run locally, ollama run gemma4:cloud runs a cloud-hosted variant through the same command. The CLI verb is unchanged — only the execution location differs.

Check your understanding

1.1 Ollama is built on which underlying project?

1.2 What does the :cloud tag (e.g. ollama run gemma4:cloud) change?