Google introduced Gemma, a new family of open LLMs designed for a wide range of applications. Gemma models are lightweight, optimized for deployment on diverse hardware, and available on Hugging Face. You can access Gemma for research and building AI-powered products.