I have to concur with you here because I'm running 3.8-flash-next in the largest unsloth Q8 version on a 256GB system for daily use, and I've recently compared it head to head vs a Q4 that I have on a 128GB system and the Q4 is just plain not as good. The Q4 is given only more basic tasks.

I can't even imaging letting a Q2 anywhere near a code base that I care about. To generate a bird on a bicycle as a test? Sure. Actual things I don't want it to screw up? nooooooooo.