KVQuant icon indicating copy to clipboard operation
KVQuant copied to clipboard
verified publisher icon SqueezeAILab

could it support LLAMA-3.1-8B-INSTRUCT?

Open LiMa-cas opened this issue 1 year ago • 1 comments

LiMa-cas avatar Dec 19 '24 09:12 LiMa-cas

你好 我也遇到了相同的问题，在运行时报错：不支持GQA，但是论文里提到支持llama3 所以很困惑

Image

Duze204 avatar Aug 10 '25 16:08 Duze204