LLaMA: Open and Efficient Foundation Language Models
Abstract
We introduce LLaMA, a collection of foundation language models ranging from 7B to 65B parameters. We train our models on trillions of tokens, and show that it is possible to train state-of-the-art models using publicly available datasets exclusively, without resorting to proprietary and inaccessible datasets. In particular, LLaMA-13B outperforms GPT-3 (175B) on most benchmarks, and LLaMA-65B is competitive with the best models, Chinchilla-70B and PaLM-540B. We release all our models to the research community.
Community
Amazing model!
Yes, but would love to get access to it :D
absolutely incredible based on all my testing so far!
So appreciative of this work!
Would love to have access hehe
It's amazing work. Better they share the weights for research than no sharing at all, but they should share it with a license which allows commercial usage, too. It would feedback educatively, because LeCun struggles to see the full potential LLMs have. Multimodality is handy, not key.
Petition to free the weights 🎉 (src tweet)
 |
 |
|---|
Meta AI, please upload them weights under more permissive license 😘
Yours truly,
AI Community
