Hardware barrier for local LLMs called 'out of hand'

A Reddit post argues that local LLMs are no longer accessible due to soaring VRAM requirements, with 8-16GB no longer sufficient. The discussion has 65 comments, reflecting community concern about the democratization of local AI.