Quantized Llama models with increased speed and a reduced memory footprintai.meta.com 508 pointsegnehots1 year ago122 commentsSaveHideCopy link On HNComments
Comments