And yes I've tried local LLMs but with just 1x RTX 5090 it's painfully slow and useless

→ View original post on X — @levelsio