logoalt Hacker News

system2today at 5:02 PM0 repliesview on HN

Until I can buy an 80GB VRAM GPU, I won't attempt to do it. A local LLM is always missing something that needs a bigger model.