.
When quantizing with Llama.cpp, use "--vocab-type bpe" as arg for quantize.py
Hello dude, I saw this message again in my mail box, did you fixed your issue?
Β· Sign up or log in to comment