meituan
/

DeepSeek-R1-Block-INT8

量化类型

#21

by WYBJ - opened Apr 11

WYBJ

Apr 11

•

量化类型是W8A8还是W8A16? 从权重格式上看像A16，因为只看到了weight_scale，如果是A8，麻烦提供/解释推理时的类型变换

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment