Built with Meta Llama 3
Discover more possibilities, learn and create with your own intelligent assistant.
Meta Llama 3
Meta Llama 3 is our most advanced model to date, capable of complex reasoning, following instructions, visualizing ideas, and solving nuanced problems.
Now available with both 8B and 70B pretrained and instruct versions to support a wide range of applications. Experience Meta Llama 3 on meta.ai .
Through research and community collaboration, we're advancing the state-of-the-art in Generative AI, Computer Vision, NLP, Infrastructure and other areas of AI. We're also applying our learnings to innovative, safer products, tools and experiences across our family of apps.
Learn more about usWe believe the responsible advancement of AI should be a shared community initiative.
AI Research
Together, we’re responsibly advancing the state-of-the-art in AI through fundamental and applied research. Our goal is apply our learnings to create products and experiences that benefit all people.
AI Resources
From research exploration to large-scale production deployments, we share our frameworks, libraries, demos, system cards, models, tools, such as PyTorch, Glow and more.
Subscribe to our newsletter to keep up with AI at Meta news, events, research breakthroughs, and more.
Sign upProduct experiences
Foundational models
Our approach
Research
Product experiences
Latest news
Foundational models