Meta’s AI translations tool that auto-dubs Instagram and Facebook Reels between English and Spanish with voice cloning and ...

CEO Mark Zuckerberg has marketed Llama as a major contender in the AI model wars, saying it's soon to be “the most advanced ...

On Friday, The Information reported that Meta was preparing to tear down its existing AI org and reorganize it into four new ...

The new structure is meant to “accelerate” the company’s pursuit of so-called superintelligence, according to an internal ...

But Meta also makes the claim that the larger-parameter-count Llama 3 model, Llama 3 70B, is competitive with flagship generative AI models, including Gemini 1.5 Pro, the latest in Google's Gemini ...

Meta restructures its AI group into four focused teams to accelerate the pursuit of superintelligence and better compete with ...

Today, Meta announced a new family of AI models, Llama 2, designed to drive apps such as OpenAI's ChatGPT, Bing Chat and other modern chatbots. Trained on a mix of publicly available data, Meta ...

In addition to its Llama 3 announcement on Thursday, Meta said that it's getting serious about Meta AI. For one, Meta AI will now live on a standalone site, where users can input queries for free.

Meta tells Fast Company that the Llama 2 models were trained on 40% more tokens (words or word parts) than the original Llama 1 models, and can read and remember far longer prompts—up to 4,000 ...

In training Code Llama, Meta used the same data set it used to train Llama 2 — a mix of publicly available sources from around the web. But it had the model “emphasize,” so to speak, the ...

Since Meta released Llama 2 as a (mostly) open-source project in July, the AI model has become a huge hit. So much so, that some experts are worried this powerful tool might be misused by bad actors.

Meta in a blog post said that the larger 405B Llama 3.1 model outperformed models such as Nemotron-4 340B Instruct, GPT-4, and Claude 3.5 Sonnet in benchmark tests such as MMLU, MATH, GSM8K, and ...