Unleash the full potential of large language models (LLMs) on your edge devices, even when there’s spotty internet. This session explores a hybrid approach that combines the power of cloud-based LLMs with the efficiency of on-device models. Learn how to intelligently route queries, enabling laptops and mobile phones to perform complex tasks while maintaining snappy performance. View demos of efficient task routing that optimizes for quality and cost to ensure your apps run smoothly, even during network disruptions.

Experience a new way to interact with LLM-powered agents! With Gemini 2.0 and Multimodal Live API, users can give audible instructions and show visual content from a camera or screen, while receiving spoken responses from the model. This enables more natural, timely communication and unlocks multimodal agent workflows. This session showcases how existing agent experiences can be adapted for voice and visual cues, and explores new possibilities with this technology.