Make: Volume 91

Courtesy of Jorvon Moss, Shawn Hymel

2. RUN THE TINYLLAMA LLM

In September 2023, AI researcher Zhang Peiyuan

started the TinyLlama project, a fully open-

source, highly-optimized LLM with only 1.1 billion

parameters. It is based on the Llama 2 model and

can generate responses quite quickly. It’s not as

accurate as the newer generation of small LLMs,

such as Llama 3, but it will run on the Pi 5.

Download the latest version of TinyLlama with

Ollama, then run an interactive shell to chat with

TinyLlama:

$ ollama pull tinyllama

$ ollama run tinyllama

You’ll be presented with a prompt. Try asking the

AI a question or have it tell you a joke (Figure

B

). ...