Welcome Llama 4 Maverick & Scout on Hugging Face

Meta has released Llama 4, featuring two variants: Maverick and Scout, available on Hugging Face. Maverick is designed for general-purpose tasks, while Scout is optimized for efficiency and lower latency, employing a modified transformer architecture that reduces parameters without sacrificing performance. This release enables practitioners to leverage state-of-the-art capabilities in LLMs while optimizing resource usage for deployment in real-world applications.