I’m trying to load an 8 bit quantized version of llama3 on my local laptop (linux) from llama.cpp, but the process is getting killed due to memory exceeding
Is there any way around this?