demobook

Sequence multi-element scenes in Gemini Omni

Demo summary

A demonstration of cause-and-effect logic where Gemini Omni generates a street market scene with wind affecting multiple objects in sequence.

Step-by-step

  1. Enter a prompt describing a single cause followed by multiple sequential effects
  2. Specify interdependent actions for different objects in the scene
  3. Generate the video to observe the causal chain

Tips

  • Test the AI's logic by sequencing multiple reactions to a single cause, such as wind affecting different materials
  • Look for the causal chain to hold even if specific secondary movements appear slightly floaty

Highlights

It gets the logic mostly right. The dog's movement is a little floaty, but the causal chain holds.

All demos from “Goodbye Seedance... Gemini Omni: Google’s New AI Video Model is INSANE

  1. 1:500:30Generate physics-based video with Gemini OmniThe user demonstrates Gemini Omni's physics understanding by prompting it to create a video of a ceramic teapot falling and shattering on a tile floor.GeminiText to Video
  2. 2:200:33Sequence multi-element scenes in Gemini OmniCurrentA demonstration of cause-and-effect logic where Gemini Omni generates a street market scene with wind affecting multiple objects in sequence.GeminiAI Realistic Image Generator
  3. 2:550:30Animate architectural photos with Gemini OmniThe user combines a reference image of a brutalist building with a text prompt to generate a time-lapse video with moving clouds and shadows.GeminiImage to Video
  4. 5:070:45Create a talking AI avatar in Gemini OmniThe video walks through the process of recording a 30-second clip to train Gemini Omni to generate a talking digital avatar of the user.GGoogle OmniAI Avatar Video Generator
  5. 6:271:22Apply visual style presets in Gemini OmniThe user demonstrates various style templates including Noir, 80s music video, Origami, and Pixelate to transform the look of generated video clips.GGoogle OmniVideo to Video
  6. 9:010:45Generate animated memes with Gemini OmniThe 'Meme Me' template is shown taking a user photo and automatically applying meme formatting, typography, and subtle facial animation.GeminiAI GIF Generator
  7. 10:000:52Conversational video editing in Gemini OmniThe user demonstrates editing a video by simply typing chat commands to change the environment from a park to a beach without regenerating the subject.GeminiAI Video Editor
  8. 10:520:43Precise object editing and lighting effectsGemini Omni is shown performing micro-edits such as changing a jacket's color and adding a light-tracking lens flare through text commands.GeminiAI Inpainting
  9. 14:540:37Create video sequences from multiple photosGemini Omni takes five separate photos and generates a coherent 10-second video sequence with smooth transitions and a consistent mood.GeminiImage to Video
  10. 15:350:41Animate static infographics with Gemini OmniThe user demonstrates feeding a flat chart into the model and prompting it to animate the data points in a logical reveal sequence.GeminiAI Animation Generator
  11. Watch “Goodbye Seedance... Gemini Omni: Google’s New AI Video Model is INSANE” →

AI Realistic Image Generator

  1. 2:200:33Sequence multi-element scenes in Gemini OmniCurrentA demonstration of cause-and-effect logic where Gemini Omni generates a street market scene with wind affecting multiple objects in sequence.AI Master