Father-in-law addicted to local LLMs admits they have no idea how to train one

multimodalart · x · 2026-07-31

A user shared an amusing anecdote about his father-in-law, a hardcore 'localmaxxer' who gets completely absorbed running models locally using llama.cpp.

While they were running Qwen3.6-27B, the user asked him what it would cost to train such a model today. The father-in-law gave a blunt and unforgettable answer: 'We can’t, we don’t know how to do it.'

Original post →

More from Fun

Fun channel →