Microsoft and Meta announced last Tuesday (18) the release of the second generation of Llama 2, a large-scale AI language model that is now available in open source. This release puts companies in a competitive position thanks to the intelligence of OpenAI, giving developers and researchers free access for research and commercial purposes.

Compared to the first version of AI, Llama 2 was a trainer with 40% more open data compared to the first. Thus, it is expected to be able to handle almost twice as much context as before.

Meta noted that compared to other open source large language models (LLMs), LLaMA 2 outperforms others in terms of coding, competence, reasoning, and performance on knowledge tests. However, the company acknowledged that LLaMA 2 still falls short of the performance of its closed-source competitors such as OpenAI’s GPT-4.

Microsoft, which has already invested heavily in OpenAI, investing just $13 million in 2023 alone, is now backing the two main AI players. The tech giant also announced that LLaMA 2 will be available on its Azure cloud platform as well as on the Windows operating system.

The meta will continue despite the criticism

Meta’s decision to make LLaMA open source was met with criticism from two U.S. senators in June, who raised concerns that the “seemingly minimal” protections in the first version of LLaMA could allow attackers to engage in “illegal activities.”

Qualcomm, one of the leaders in telecommunications, also said it is partnering with Meta to expand LLaMA’s presence on more devices. The company said in a statement that LLaMA 2 will be integrated into Snapdragon-powered smartphones and PCs starting in 2024.

Mundo Conectado Deal Center: Selection of Discounts and Lowest Prices
Best deals on electronics, cell phones, TVs, soundbars, drones and more

Source: Microsoft, Meta.

…..