Who is this for?

This is an introductory topic for anyone interested in running the Llama 3 model on a Raspberry Pi 5, and learning about techniques for running large language models (LLMs) in an embedded environment.

What will you learn?

Upon completion of this Learning Path, you will be able to:

  • Use Docker to run Raspberry Pi OS on an Arm Linux server.
  • Compile a Large Language Model (LLM) using ExecuTorch.
  • Deploy the Llama 3 model on an edge device.
  • Describe how to run Llama 3 on a Raspberry Pi 5 using ExecuTorch.
  • Describe techniques for running large language models in an embedded environment.

Prerequisites

Before starting, you will need the following:

Summary

AI-assisted

This summary was drafted with an approved AI-assisted workflow and reviewed by Arm contributors before publication. Human technical review remains part of the process so the final page reflects engineering rigor, accuracy, and Arm editorial standards.

Close
?
You’ll build and deploy Llama 3 on a Raspberry Pi 5 with ExecuTorch. First, you’ll use a Docker container running Raspberry Pi OS to create a Python environment, install dependencies, and compile a quantized model for edge execution. Then, you’ll prepare 64-bit Raspberry Pi OS, transfer the artifacts, and run prompts to verify on-device responses.

Frequently asked questions

AI-assisted

These FAQs were drafted with an approved AI-assisted workflow and reviewed by Arm contributors before publication. Human technical review remains part of the process so the final page reflects engineering rigor, accuracy, and Arm editorial standards.

Close
?
Which Raspberry Pi OS image should I install on the Raspberry Pi 5?
Install the 64-bit version of Raspberry Pi OS. Use Raspberry Pi Imager to prepare the SD card.
Should I set up ExecuTorch on the development host or directly on the Raspberry Pi?
Set up ExecuTorch inside the Raspberry Pi OS shell running in the Docker container on your Arm Linux machine. Doing so isolates dependencies and prepares binaries for deployment to the Raspberry Pi 5.
Do I need to quantize the Llama 3 model for the Raspberry Pi 5?
Quantization is often used to reduce the memory footprint of large models for memory-constrained devices. Choose an approach that fits your device constraints.
How can I experiment with the model on the Raspberry Pi 5?
Run llama_main with a prompt, then try different prompts and command-line options. You can adjust the prompt and the number of CPU threads to explore the model’s behavior.
Are there usage restrictions for Llama 3?
Yes. Llama models are subject to an acceptable use policy and a responsible use guide. Review those materials before using or distributing the model.
Next