What is Meta's new LLaMA?

a state-of-the-art foundational LLM

Photo by Dima Solomin on Unsplash

Introducing Large Language Model Meta AI (LLaMA)

Hey Everyone,

Meta has BlenderBot, Galactica, and many other experiments in LLMs and around A.I. MetaAI is a decent R&D lab team. It’s not always evident at least to me, what their achievements in A.I. are however.

Most recently they have released a new open-source high performance LLM from Meta (formerly FAIR) called LLaMA.

They say this work is a result of using smaller models trained on more tokens. Meta AI researchers trained LLaMA-65B and LLaMA-33B on 1.4 trillion tokens, and their smallest model, LLaMA-13B on one trillion tokens — more than 3x a similar public model.

LLaMA (Large Language Model Meta AI) achieves results competitive with the best currently released models while being smaller & more efficient — increasing accessibility to this technology for even more researchers working on this important subfield of AI across the globe.

Meta AI seems to think it’s some kind of a good samaritan.