Awesome, thanks
@bartowski
quick question; is there any functional difference between this repo published under your handle, and the one published under the lmstudio-community handle?
Can someone confirm this model is better than qwq32b in real-world use cases, or just in benchmarks? I am about to find out with my usual test questions...
the ones on lmstudio-community are static (no imatrix), some people prefer the purity of that, i prefer the imatrix, so I release them under my own repo always with imatrix, and then lmstudio has the static ones :)
@urtuuuu really hard to say, their benchmarks certainly seem to think so, but real world use cases are much harder to judge, would love to hear your findings!
It seems at least on same level as qwq32b. Most importantly thinking time definitely much less. And I'm using Q3_km. Also i didn't expect it to get this right on first try:
Write a Python program that shows a ball bouncing inside a spinning hexagon. The ball should be affected by gravity and friction, and it must bounce off the rotating walls realistically.