The Llama 4 collection features natively multimodal AI models that utilize a mixture-of-experts architecture, enabling superior performance in both text and image comprehension. With models like Llama 4 Scout and Llama 4 Maverick, each boasting 17 billion parameters, these tools are designed for commercial and research applications, supporting tasks such as visual reasoning, natural language generation, and image captioning.