LLaMA: INT8 edition

https://github.com/tloen/llama-int8

https://github.com/facebookresearch/llama/issues/79#issuecomment-1454687232

 

This is a fork of the LLaMA code that runs LLaMA-13B comfortably within 24 GiB of RAM

 

https://github.com/go-noah/llama

LLaMA 7B example