Meta is making its large language model, Llama 2, available for free to companies and researchers, offering an alternative to costly proprietary models from OpenAI and Google. Starting today, Llama 2 is available on Microsoft's Azure cloud computing service and is optimized to run locally on Windows. Meta named Microsoft as its preferred partner for Llama 2. - The LLM is available through Amazon Web Services (AWS), Hugging Face, and other providers too.
- While Meta won't charge for using Llama 2 directly, it plans to offer the program as a paid hosted service through cloud computing companies.
- Unlike other closed-source language models, Meta has published Llama's starting and code weights.
- Opening up access to AI models means developers and researchers can stress test them and pinpoint problems, according to a Meta blog post.
Meta aims to disrupt the generative AI software market by offering free and widespread access to Llama, targeting the early dominance of OpenAI and other competitors. - After Meta released the first version of Llama in February, it received over 100,000 research requests.
- Llama 2 was trained on 40% more data than its predecessor, with over 1 million human annotations to enhance output quality.
|
