My Qwen3.8-27B task-aware quant reaches 99% of BF16 reasoning performance at 15% of the size.
My Qwen3.8-27B task-aware quant reaches 99% of BF16 reasoning performance at 15% of the size.
Deep Dive
TL;DR My TAK quant of Qwen 3.8 27b scored 82.81% on reasoning, comparted with 77.34% for the byte matched Unsloth UD IQ2_S and 83.59% for BF16. Edit: Some of you have tried coding with this reasoning-specialized quant and encountered repetition loops. Coding is outside its intended domain, but I’ll