I benchmarked quants of Qwen 3 .6b from q2-q8, here's the results:
r/LocalLLaMA
•
Open Source AI
Submitted by /u/PraxisOG [link] [comments]