Running Small LLMs (oxymoron) Locally With an Integrated GPU on a Laptop.

This is what running local models on a consumer-grade laptop without a discrete GPU looks like... Still acceptable for many use cases in my opinion, but I definitely want to upgrade to a machine with a dedicated GPU. 
_________
Specs:

Framework 13 

OS: Linux 

Hardware: AMD Ryzen 7040Series AMD® Ryzen 7 7840u w/ radeon 780m graphics × 16
__________

Run in the terminal:
ollama run --verbose  llama3.2:1b

Then run your prompt, and after the model provides its output, the statistics will display.

The "verbose" specification provides the statistics output.