Until today, open source large language models have mostly trailed behind their closed counterparts when it comes to capabilities and performance. Now, we’re ushering in a new era with open source leading the way. We’re publicly releasing Meta Llama 3.1 405B, which we believe is the world’s largest and most capable openly available foundation model. With more than 300 million total downloads of all Llama versions to date, we’re just getting started.
Llama 3.1 405B is the first openly available model with state-of-the-art capabilities in general knowledge, steerability, math, tool use, and multilingual translation. This release aims to drive innovation by supporting applications like synthetic data generation and model distillation, unprecedented at this scale in open-source models. Alongside the 405B, upgraded 8B and 70B models now offer multilingual capabilities, 128K context length, advanced reasoning, and tool use for tasks like text summarization, conversational agents, and coding assistants. The models are available for download and development on Meta’s site and Hugging Face.
For this release, we evaluated performance on over 150 benchmark datasets that span a wide range of languages. In addition, we performed extensive human evaluations that compare Llama 3.1 with competing models in real-world scenarios. Our experimental evaluation suggests that our flagship model is competitive with leading foundation models across a range of tasks, including GPT-4, GPT-4o, and Claude 3.5 Sonnet. Additionally, our smaller models are competitive with closed and open models that have a similar number of parameters.
