News for hackers. Stuff that matters.

A comment on LLaMA-7B in Pure C++ with full Apple Silicon support.

Re:LLaMA-7B in Pure C++ with full Apple Silicon support by gh2k on Sunday March 12, 2023 @09:22AM (#35119765)
Does anyone know if it's possible to split this model across multiple (identical, sli) GPUs, to run a version bigger than fits in the RAM of a single GPU?
[ Parent | on HN ]

There are no comments on this comment yet.