so the speed of prosparse-llama-2-7b on A100 GPU with powerinfer is just 16 tokens/s?? I want to know how this was tested, please.
· Sign up or log in to comment
