My Qwen3.8-27B task-aware quant reaches 99% of BF16 reasoning performance at 15% of the size.
英文标题：My Qwen3.8-27B task-aware quant reaches 99% of BF16 reasoning performance at 15% of the size.
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
打开原文相关 4 · 机会 0
中文
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
中文全文翻译正在补齐，可先查看下方英文原文。
English / 原文
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