The YouTube video is titled "Best local model for you".
It recommends the best one according to GPU specifications, and there are quite a few I haven't seen before.
But it seems to be motivated by the top predator (the part that eats the most RAM..).
OpenLLM is ranked 3rd. Above that are GLM and Kimi, which we all know very well...
My heart is full. Of course, I don't have the capacity to use them, so I can't use them,
but it's really cool to see them rise like an underdog.
I hope for good results.
▶ Original source: https://www.youtube.com/watch?v=T8v8Rxr8rMM
