Running Your First Model

In this guide, you'll set up and run your first Large Language Model (LLM) on your local machine using Ollama. We’ll cover prerequisites, compare local versus cloud deployments, walk through Ollama’s setup process, and demonstrate how to chat with a model—all offline and without usage fees.

Prerequisites

Before you begin, make sure you have:

• The Ollama app installed on your computer
• Access to the Ollama CLI (ollama command)

Local vs. Cloud Deployment

You have two options for running LLMs:

Ollama Setup Process

Ollama automates:

1. Downloading the model files
2. Installing dependencies and preparing the environment

Running a Model with Ollama

We’ll start with LLaMA 3.2, Meta’s open-source LLM. To launch:

ollama run llama3.2

If the model isn’t already on your system, Ollama will:

1. Pull the manifest and layers
2. Verify the SHA-256 digest
3. Cache metadata for faster future starts
4. Launch the model

Example output:

$ ollama run llama3.2
pulling manifest
pulling 633fc5be925f... 100% 2.2 GB
pulling fa8235e5b48f... 100% 1.1 KB
pulling 542b217f179c... 100%
verifying digest
writing manifest
✔ model llama3.2 ready

Chatting with Your Model

Once loaded, Ollama drops you into an interactive chat:

$ ollama run llama3.2
>>> hey! how are you?
I'm just a language model, so I don't have emotions or feelings like humans do, but thank you for asking! How can I help you today? Is there something on your mind that you'd like to chat about or ask for assistance with? I'm all ears (or rather, all text).
>>> /bye

Type /bye to close the session. You now have a fully offline LLM chat interface.

Next Steps

In the next lesson, we’ll:

• Explore other models supported by Ollama
• Read model descriptions and metadata
• Run additional LLMs locally

Links and References

• Ollama Documentation
• Meta LLaMA 3 Release Notes
• Local LLM Best Practices