DavidAU

/

Llama-3.3-70B-Instruct-How-To-Run-on-Low-BPW-IQ1_S-IQ1_M-at-maximum-speed-quality

like 2

Hi David,

I was wondering what the "-c 2048 -ngl 99" does at the end of the model. Not necessarily because I want to run the model from the guide, but I am interested in running models directly in llama cpp from now on. Is there any guide you would recommend? Or should I just use Kobold cpp instead and hope that its always quickly updated?

Thanks!