Meta, Microsoft team up to offer AI model Llama for commercial use
Meta Platforms announced on Tuesday that it is releasing a commercial version of its open-source artificial intelligence model Llama. This move provides startups and businesses with a free alternative to costly proprietary models sold by OpenAI and Google. The new version, Llama 2, will be distributed by Microsoft through its Azure cloud service and will run on the Windows operating system. In a blog post, Meta referred to Microsoft as their “preferred partner” for the release. Llama 2 will also be available for direct download and through other providers like Amazon Web Services and Hugging Face. The model was previously only accessible to select academics for research purposes.
Meta’s CEO, Mark Zuckerberg, stated that open source drives innovation by allowing more developers to build with new technology. He believes that making a sophisticated model like Llama widely available and free for businesses threatens the dominance of players like OpenAI in the generative AI software market. Llama was already competitive with models like OpenAI’s ChatGPT and Google’s Bard chatbot. The new version has been trained on 40% more data than its predecessor and has undergone more than 1 million annotations by humans to enhance its output quality.
Amjad Masad, CEO at software developer platform Replit, mentioned that any incremental improvement in open-source models eats into the market share of closed-source models. The availability of powerful open-source systems at a low cost reduces dependency on costly proprietary models.
This announcement follows similar plans by Microsoft’s rivals, Google and Amazon, to offer a variety of AI models to business customers. Meta aims to foster an open-source ecosystem of AI technology built using its models, which could impede rivals’ revenue streams from their proprietary technology.
However, releasing Llama into the wild also carries risks. It allows unscrupulous actors to easily build products without safety controls. In April, Stanford researchers took down a chatbot they had built using the first version of Llama after it generated problematic text. Meta believes that public releases of technology actually reduce safety risks by utilizing the collective knowledge of the crowd to identify and address problems. They have implemented an “acceptable use” policy for commercial Llama, prohibiting certain use cases including violence, terrorism, and child exploitation.
FAQs:
Q: What is Llama?
A: Llama is an open-source artificial intelligence model provided by Meta Platforms.
Q: What is the new version of Llama called?
A: The new version is called Llama 2.
Q: How will Llama 2 be distributed?
A: Llama 2 will be distributed by Microsoft through its Azure cloud service and will run on the Windows operating system.
Q: Will Llama 2 be available for download from other platforms?
A: Yes, Llama 2 will be available for direct download and through providers like Amazon Web Services and Hugging Face.
Q: How does Llama 2 compare to other models like ChatGPT and Bard?
A: Llama 2 is already competitive with models like OpenAI’s ChatGPT and Google’s Bard, and it has been trained on more data with enhanced output quality.
Q: Why is Meta releasing Llama as open source?
A: Meta believes that open source drives innovation by allowing more developers to build with new technology.
Q: What potential risks come with the release of Llama?
A: Releasing Llama into the wild may enable unscrupulous actors to build products without safety controls. However, Meta believes that public releases actually help identify and address problems through crowd feedback.
Q: Are there any restrictions on the use of Llama for commercial purposes?
A: Yes, Meta has an “acceptable use” policy for commercial Llama, prohibiting certain use cases including violence, terrorism, and child exploitation.
Meta and Microsoft collaborate to offer commercial usage of the AI model, Llama.
Meta Platforms, the parent company of Facebook, has announced the commercial release of its open-source artificial intelligence model, Llama. The model, known as Llama 2, will be distributed through Microsoft’s Azure cloud service and will be compatible with the Windows operating system. Meta referred to Microsoft as its “preferred partner” for the release. Llama 2 was previously only available to select academics for research purposes, but will now be accessible to start-ups and businesses via direct download or through providers such as Amazon Web Services and Hugging Face. The release of Llama presents a cost-effective alternative to proprietary AI models offered by companies such as OpenAI and Google. With Llama’s wider availability, the dominance of established players in the generative AI software market could be disrupted. Llama 2 has been trained on 40% more data than its predecessor and includes over 1 million annotations to enhance the quality of its outputs. The commercial release of Llama marks Meta’s pursuit of an open-source ecosystem for AI technology, where the company can benefit from infrastructure cost reductions and the creation of new consumer-facing tools by crowd-sourcing innovations. However, releasing Llama also carries risks in terms of potential misuse by unscrupulous actors. Meta has implemented an acceptable use policy to prevent certain prohibited use cases such as violence, terrorism, and child exploitation.