At the Google I/O 2026 event, Google unveiled Gemini Omni, a groundbreaking multimodal AI model that merges text, images, audio, and video into cohesive video outputs. This innovative tool allows users to upload various media types, including photos, drawings, and existing videos, and combine them into a single video. A notable feature of Gemini Omni is its conversational editing capability, enabling users to describe desired changes in simple language, such as 'turn a mirror into liquid' or 'change the background to a futuristic city.' This user-friendly approach aims to make video creation more accessible to everyone. The first iteration, named Gemini Omni Flash, is currently being rolled out through platforms like the Gemini app, Google Flow, and YouTube Shorts. Additionally, every video produced by Gemini Omni is embedded with an invisible SynthID watermark for verification purposes. While the initial version supports only voice references for audio input, Google has plans to expand audio capabilities in future updates, further enhancing the platform's versatility.