Step-by-Step Guide to Install LLaMA 3 on Windows 11 PC

Llama 3 represents Meta’s most recent advancement in large language models, ideal for a wide range of applications including answering questions, assisting with academic assignments, and much more. By setting up Llama 3 on your Windows 11 device, you can access it anytime, even without internet connectivity. This guide will demonstrate how to set up Llama 3 on your Windows 11 computer.

Installing Llama 3 on a Windows 11 Computer

The process of installing Llama 3 on your Windows 11 device using Python necessitates a certain level of technical proficiency. Nevertheless, alternative methods exist that simplify local deployment of Llama 3. I will outline these techniques.

To install Llama 3, you will need to run specific commands in the Command Prompt. Note that this will only grant access to the command line version; additional steps are required for utilizing its web interface. Both processes will be covered here.

Setting Up Llama 3 on Windows 11 via CMD

First, you must install Ollama on your Windows computer to deploy Llama 3. Follow these steps:

1. Navigate to the official Ollama website.
2. Select the Download option, then choose Windows.
3. Click on Download for Windows to save the executable file to your computer.
4. Execute the downloaded exe file to install Ollama on your device.

After Ollama is successfully installed, restart your computer. It should be running in the background, visible in the System Tray. Then, visit the Models section on the Ollama website to view available models.

The Llama 3.1 model is offered in three configurations:

• 8B
• 70B
• 405B

The 405B configuration is the most demanding and may not function on a lower-end machine. Llama 3.2 provides two options:

• 1B
• 3B

Choose a Llama version for installation—if opting for Llama 3.2, click on it. In the ensuing drop-down menu, select your desired configuration. Next, copy the command displayed next to it and paste it into the Command Prompt.

Here are the commands for the Llama 3.2 model:

ollama run llama3.2:3b

To install the Llama 3.2 1B configuration, enter:

ollama run llama3.2:1b

Open the Command Prompt, type one of the above commands according to your needs, and press Enter. The download process will take a little while, depending on your internet connection. Upon completion, a success message will appear in the Command Prompt.

You can then type your input to interact with the Llama 3.2 model. For installing the Llama 3.1 model, use the commands available on the Ollama website.

The next time you open Command Prompt, you can utilize the same command to run Llama 3.1 or 3.2.

One limitation of installing Llama 3 via CMD is the lack of saved chat history. However, deploying it through a local host allows your chat history to be saved, in addition to providing an improved User Interface. The following section covers how to achieve this.

Deploying Llama 3 with a Web UI on Windows 11

Utilizing Llama 3 through a web browser not only enhances user experience but also preserves chat history, a feature absent when using CMD. Here’s how to get Llama 3 running in your web browser.

To access Llama 3 via a web browser, ensure both Llama 3 through Ollama and Docker are set up on your system. If you haven’t installed Llama 3, proceed with the Ollama installation as detailed earlier. Next, download and install Docker from its official website.

After installing Docker, open it and complete the sign-up process to create an account, as Docker will not initiate without it. Upon signing in, minimize Docker to the System Tray; ensure both Docker and Ollama are active in the background to use Llama 3 through your web browser.

Open the Command Prompt, copy the following command, and paste it:

docker run -d -p 3000:8080 --add-host=host.docker.internal:host-gateway -v open-webui:/app/backend/data --name open-webui --restart always ghcr.io/open-webui/open-webui:main

This command will take some time to download the necessary files, so be patient. Once it completes, open Docker and navigate to the Containers section on the left side. You should see a container created automatically for port 3000:8080.

Click on the port 3000:8080, which will launch a new tab in your default web browser. You may need to sign up and log in to use Llama 3 via the web browser. If you check the address bar, it will show localhost:3000, indicating that Llama 3 is hosted locally, allowing use without internet access.

Select your preferred Llama chat model from the drop-down menu. To incorporate additional Llama 3 chat models, install them via Ollama using the corresponding commands; they will then be available in your browser.

Your chat history will be saved and retrievable on the left side. When you’re done, log out of your session in the web browser, then open Docker and hit the Stop button to shut it down before closing Docker.

The next time you wish to access Llama 3 in your web browser, start both Ollama and Docker, wait a few minutes, then click on the port in the Docker container to launch the localhost server. After signing in, you can commence using Llama 3.

I hope this information proves helpful.

Can Llama 3 operate on Windows?

Your ability to run Llama 3 on your machine depends on its hardware specifications. The lightest version, the 1B model, can be installed and operated through the command prompt.

What amount of RAM is required for Llama 3?

To run the Llama 3.2 1B model, your system should be equipped with at least 16 GB of RAM, along with a robust GPU. Higher variants of Llama 3 will demand even more resources from your system.

Source