First Principles
Avsnitt

EP64 - Small Language Models: How 100x Less AI Stays Just as Smart

Dela

Ask your phone a question in airplane mode and it answers instantly, with no internet, no server farm, no cloud involved — yet the model inside is a fraction of the size of the ones that power web chatbots. How can an intelligence a hundred times smaller still think? We open the black box of distillation and quantization to reveal that being smart was never really about size.

Podden och tillhörande omslagsbild på den här sidan tillhör First Principles Podcast. Innehållet i podden är skapat av First Principles Podcast och inte av, eller tillsammans med, Poddtoppen.