Changkang Li

Results 1 issues of Changkang Li

cogvlm-chat-v1.1 model H800 machine Why is it that running inference on two cards is slower than running on a single card?"