Gemini is Google's most capable model. It is built from the ground up for multimodality, seamlessly combining and understanding text, code, images, audio, and video. We begin the session with a coding demo of Gemini's multimodal capabilities using Google Cloud Vertex AI. We also demonstrate building agents using Gemini. We demonstrate context-switching and tool-calling capabilities. We proceed with a demo of Gemini's ability to process large inputs. We conclude the session by building a Multimodal RAG application using Gemini.