Comment by metaskills

A small experiment to see if we are there yet with highly virtualized CPU compute and Small Language Models (SLM). The answer is a resounding maybe, but most likely not. Huge thanks to Justine for her work on Llamafile supported by Mozilla. Hope folks find this R&D useful.