Meta Releases Commercial Version of Open-Source AI Model 'Llama'


Meta (formerly Facebook) has introduced a commercial version of its open-source artificial intelligence (AI) model, Llama, called Llama 2. The model will be accessible through Microsoft's Azure cloud service and is compatible with the Windows operating system. Microsoft has been selected as Meta's preferred partner for the release.

Previously, Llama was exclusively available to select academics for research purposes. However, with the commercial release, Llama 2 can be downloaded directly and accessed through platforms like Amazon Web Services and Hugging Face. The aim is to provide startups and businesses with a free alternative to expensive proprietary AI models offered by companies like OpenAI and Google.

Mark Zuckerberg, CEO of Meta, emphasized the importance of open-source technology in driving innovation by enabling more developers to leverage new technologies. The availability of Llama to a wider audience poses a threat to the dominance of proprietary generative AI software offered by players like OpenAI, which is backed by Microsoft.

Llama 2 has been trained on a larger dataset than its predecessor, incorporating over 40 percent more data. Additionally, more than 1 million human annotations have been utilized to enhance the quality of its outputs. The model is designed to compete with models such as OpenAI's ChatGPT and Google's Bard chatbot.

Meta's strategy of offering a sophisticated open-source AI model like Llama for free aims to disrupt the market share of closed-source models. By promoting the use of equally powerful open-source systems, businesses can benefit from cost-effective implementation and reduce reliance on proprietary solutions.This move aligns with similar strategies adopted by major cloud providers like Google and Amazon, who are also offering a range of AI models to their business customers.

By fostering an open-source ecosystem built on Meta's models, the company aims to undermine revenue streams associated with proprietary technology. This approach empowers developers and reduces the value of closed-source models offered by competitors.

While the release of Llama brings numerous advantages, including crowd-sourced advancements and bug fixes, it also poses risks. There is a concern that unscrupulous actors may misuse the technology. To mitigate this, Meta has implemented an "acceptable use" policy for commercial Llama, prohibiting specific use cases involving violence, terrorism, child exploitation, and other criminal activities.

Overall, Meta believes that public releases of technologies enhance safety by harnessing collective wisdom to identify and address potential issues. Additionally, it allows the company to benefit from industry-standard tools and infrastructure.