– Meta, formerly known as Facebook, announces new large language model LLaMA available to researchers and affiliated entities.

– LLaMA has the potential to outperform competitors like GPT-3 and Google’s Chinchilla70B and PaLM-540B.

– Meta’s LLaMA model boasts cleaner data and architectural improvements contributing to its enhanced training stability.

Meta, formerly known as Facebook, has joined the AI arms race with the release of a new large language model called LLaMA, which stands for Large Language Model Meta AI. While users cannot currently talk to the model like ChatGPT or Bing AI chatbot, LLaMA will be available to researchers and entities affiliated with the government, civil society, and academia under a non-commercial license.

Large language models are a type of artificial intelligence that can process and understand natural language text. LLaMA is a promising addition to the field and has the potential to outperform its competitors, according to Meta. The company claims that LLaMA’s 13 billion parameter version can even outperform GPT-3, a predecessor to the model on which ChatGPT is built.

Furthermore, Meta describes its 65 billion parameter LLaMA model as competitive with Google’s Chinchilla70B and PaLM-540B, which are even larger than the model used by Google to demonstrate its Bard chat-powered search. A Meta spokeswoman attributed the performance to a larger quantity of “cleaner” data and “architectural improvements” in the model that enhanced training stability.

This announcement comes as large language models have become increasingly popular in recent months, with companies like OpenAI, Microsoft, and Google developing some of the largest and most powerful models in the world. LLMs have shown promise in generating text, having conversations, summarizing written material, and more complicated tasks like solving math theorems or predicting protein structures, as CEO Mark Zuckerberg wrote on Friday.

Meta released a large language model OPT-175B in May of last year, which formed the basis of a new iteration of its chatbot BlenderBot. Later, the company introduced a model called Galactica, which could write scientific articles and solve math problems. However, the company quickly pulled down the demo after it generated authoritative-sounding false responses.

While LLaMA is not yet available for commercial use, it is expected to make waves in the AI industry once it is. Meta’s latest release shows the company’s commitment to advancing AI technology and keeping up with its competitors.

———

#Meta #LLaMA #AI #LanguageModel #GPT3 #GoogleChinchilla70B #PaLM540B