Llama 4

// about Llama 4

Llama 4 is Meta's latest generation of open-weight large language models, featuring a Mixture-of-Experts architecture that achieves frontier-level capability while remaining freely downloadable under Meta's community licence. The family spans Scout (17B active parameters, 10M-token context), Maverick (17B active, multimodal), and Behemoth (288B active, teacher model) variants. Its open nature makes it the most widely self-hosted and fine-tuned LLM family, deployed in production across research labs, enterprises, and consumer applications worldwide.