Meta has stepped up its game and wants everyone to use its artificial intelligence. The company announced this week that LLaMA 2, the new generation of its language model, will be completely free for both research and commercial use.
In this way, Meta aims to close the gap with OpenAI and its popular AI, GPT-4, which is free to use and is implemented in both ChatGPT and the new Microsoft Bing.
In its statement, Meta says that it has decided to make the second version of LLaMA open source in order to provide access to AI tools for companies, startups, and researchers, facilitating community experimentation.
According to Meta, LLaMA 2 was launched with 40% more data than the first version of LLaMA, including “publicly available data sources on the internet.”
Furthermore, by opening LLaMA 2 to everyone, Meta aims to improve key aspects of its AI, such as transparency and security. The new generation of the AI has undergone a process known as “red teaming,” where “contradictory inputs were generated to fine-tune the model,” both internally and externally.
We believe that an open approach is suitable for the development of current AI models, especially in the generative space, where technology is advancing rapidly,” stated Meta in their announcement.
“Opening access to current AI models means that a generation of developers and researchers can subject them to stress tests, identifying and resolving issues quickly, as a community.
With this announcement, LLaMA 2 will be available on Azure, Microsoft’s cloud platform. Additionally, Meta has also confirmed the availability of LLaMA, the first generation of the language model, on providers such as Amazon Web Services and Hugging Face, among others.
Some of the links added in the article are part of affiliate campaigns and may represent benefits for Softonic.
