Please point out where you took that quote from, or what statement I made that could even be remotely paraphrased to that, or stop accusing people of saying things that you've misunderstood
> Not everyone bought a GPU to run LLMs on. That is quite a bit of VRAM, especially for non-desktops