Meta, Social media company, on Tuesday, announced the general availability of Llama 2, the next generation of its open-source large language model (LLM).
Meta said it has partnered with Microsoft for Llama 2 is the follow-up to Llama. Unlike Llama, Llama 2 is available free of charge for research and commercial use. It also includes model weights and starting code for the pre-trained model and conversational fine-tuned versions too.
Llama 2 is more powerful and efficient than its successor. According to Meta, Llama 2 chat models were trained on publicly available online data sources and over 1 million human annotations. It has double the context length of Llama 1.
Llama 2 is available for downloading in the Azure AI model catalogue, Amazon Web Services (AWS), and Hugging Face, among other providers. It is also optimized to run locally on Windows.
Meta advocated for the open approach in the development of today’s AI models saying it will benefit society.
“By making AI models available openly, they can benefit everyone. Giving businesses, startups, entrepreneurs, and researchers access to tools developed at a scale that would be challenging to build themselves, backed by computing power they might not otherwise access, will open up a world of opportunities for them to experiment, innovate in exciting ways, and ultimately benefit from economically and socially,” it said.
LLaMA 2, like all LLMs, suffers from the same issues that afflict its counterparts—tendencies to generate falsehoods and offensive language.
Meta said, “Opening access to today’s AI models means a generation of developers and researchers can stress test them, identifying and solving problems fast, as a community. By seeing how these tools are used by others, our own teams can learn from them, improve those tools, and fix vulnerabilities.”
With the advancement of Llama, Meta takes on its rivals OpenAI and Google in the race to develop the most efficient and widely used LLM. So far, OpenAI’s ChatGPT takes the lead.