Meta releases Llama 2, next generation of its open source large language model: All you need to know

The service has been made available for use and comes free-of-cost.

Meta Llama 2 release live updates: Meta on Tuesday launched Llama 2 which it says is the ‘next generation of our open source large language model (LLM).’ The Mark Zuckerberg-helmed company has also made Llama 2 – a potential rival to both ChatGPT (OpenAI) and Bard (Google) – available for research and commercial purposes, and the service can be used free-of-cost.

What is Llama 2?

Representational Image (Reuters)

According to Hugging Face, which is among Meta's global partners and supporters on the project, Llama 2 is a ‘family of state-of-the-art open-access language models' released by the tech giant.

Llama 2: Features

(1.) As per Meta, Llama 2 received training on 40% more data than its predecessor, Llama 1, and has double the context length. The model is available in three sizes: 7B, 13B, and 70B.

(2.) Facebook's parent organisation also claims that the new LLM ‘outperforms’ other open source LLMs on many external benchmarks, including reasoning, coding, proficiency, and, knowledge tests.

(3.) Pre-trained on publicly available online data sources, the Llama-2-chat, a ‘fine-tuned’ model, leverages publicly available instruction datasets, and more than 1 million human annotations.

Global partners and supporters on Llama 2

It is available for use immediately through respective providers. Available in the Azure AI model catalog, it is also optimised to run locally on Windows.