Meta AI has announced the release of Llama 2, the latest update of its large-scale language artificial intelligence (AI) model. July 18th.
company Said In an additional Twitter comment:
“We believe an open approach with broad accessibility is appropriate for the development of Al models today. Today we release Llama 2…for research and commercial use. is available free of charge to
The new release surpasses Llama 1 in many ways. Llama 2 is trained on 40% more data than its predecessor, has a longer context length, uses models with up to 70 billion parameters, and features a pre-trained model trained on 2 trillion tokens .
The use of public data is also a focus of the announcement. Llama 2 was pre-trained based on publicly available online data. Its fine-tuned model, Llama-2-chat, utilizes a public instruction dataset and over 1 million human annotations.
Meta’s announcement provided data showing that Llama 2 “outperforms” competing open-source language models when tested against several benchmarks, including coding, reasoning, proficiency, and knowledge tests. I’m here. In the foreword to the accompanying paper, Meta notes that at the moment, other publicly released language models are completely competitive with closed models such as OpenAI’s GPT, Google’s Bard, Anthropic’s Claude, etc. I argue that it is not possible.
Meta also released a statement declaring that it intends to take a responsible approach to AI. It argues that “visibility, scrutiny, trust” and an open approach are beneficial. The statement has been signed by dozens of industry insiders.
Rama 2 is part of the big picture
This announcement represents the latest step towards the launch of Llama. Meta released Llama under a non-commercial license in his February. Shortly after its limited release, Llama’s source code was leaked to the public in full.
However, with the latest release, the project’s code can now be used in commercial applications. This means that AI models may soon be used in a wide variety of products.
Meta has named several major companies as partners and supporters, including Microsoft, AWS, IBM, and Nvidia. Crypto-friendly VC firm Andreessen Horowitz (a16z) is one of its backers. These and other companies will build on Llama, provide access through cloud services, and conduct research, Meta said.
Meta previously announced plans to launch a consumer AI app that will be integrated with the Facebook platform and other products. We are also launching voice generation tools and advertising suites that utilize AI.
Meta’s AI offerings are not cryptocurrency or blockchain-agnostic, but more popular expectations about AI have created a robust blockchain AI sector. AI-related cryptocurrency tokens currently have a market capitalization of at least $2.67 billion.
