Meta has unveiled its latest generation of open-source large language model, Llama 3, with two text-based models trained with 8B and 70B parameters. These models offer improved reasoning capabilities and are integrated into the Meta AI assistant, positioning them as best-in-class open source models. The company aims to continue driving innovation in AI across various applications, tools, and optimizations.
Initially anticipated for May, the release of Llama 3 was delayed due to unforeseen circumstances. However, the latest models have been evaluated against other industry benchmarks and showcase their superior performance. They have undergone rigorous filtering processes to ensure data quality and reliability. Additionally, more than 5% of the pre-training data set includes information in languages other than English, preparing for upcoming multilingual use cases.
To promote safe interactions with Llama 3, Meta has introduced resources like Llama Guard 2, Code Shield, and CyberSecEval 2. The latest Llama 3 models have been integrated into the Meta AI assistant across various platforms like Facebook, Instagram, WhatsApp, and Messenger. Users can download the models and soon access them on various platforms including Amazon Web Services, Google Cloud, IBM WatsonX