Are these done as LLaMA deltas still? I.e. do I need to apply a patch to LLaMA, ... | Hacker News

Hacker Newsnew | past | comments | ask | show | jobs | submit

		quickthrower2 on June 13, 2023 \| parent \| context \| favorite \| on: Llama.cpp: Full CUDA GPU Acceleration Are these done as LLaMA deltas still? I.e. do I need to apply a patch to LLaMA, and so I still need to source LLaMA?

Tostino on June 13, 2023 [–]

Most of them are merged models, so you don't need the base model.

It's stupidly simple to get going.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact