Hao Zhang
Hao Zhang
@RedmiS22018 looks great. Do you mind submitting and end-to-end PR for this feature? Thanks
could you invite me to the wechat group and also assign me an admin? thanks
I'll update the README with an official wechat group page soon.
I'll test your dockerfile this week
@bradfox2 Regarding GPTQ, is the performance degeneration specific to T5 or to all LLMs?
@ss-zheng thanks, will merge soon.
@andy-yang-1 please try and verify the PR
@DachengLi1 is looking into it
pending test by @ZYHowell
@alanxmay this is just a workaround. Most of our users indeed used this workaround.