Supports text, image, and video reference workflows.
Multimodal video model
Gemini Omni Video Generator
Generate multimodal videos from text, images, or video references with Gemini Omni Video.
Generated videos will appear here.
Session historyVideos generated in this session will appear here.
What Gemini Omni Video is good for
Use the model page as a clean entry point for generation.
Offers 720p, 1080p, and 4K options.
Useful for cinematic and storyboard-guided video generation.
Common workflows
Storyboard videosCampaign clipsReference-driven shots
Model settings at a glance
Quick facts for choosing the right model before you generate.
Input modes
Text, image, or video to video
Output options
720p, 1080p, or 4K, 4 / 6 / 8 / 10 seconds
Best fit
Storyboard-guided shots and multimodal reference workflows
Gemini Omni Video prompt examples
Use these examples as starting structures, then adjust subject, style, camera, lighting, and output format.
A cinematic product reveal for a wearable device, slow dolly in, soft reflections, clean tech studio, elegant final frame.
Transform the uploaded reference into a fashion campaign video with controlled camera movement, realistic fabric motion, premium lighting.
Generate a vertical social ad showing a drink can opening with condensation, fresh fruit splash, bright summer color grade.
Use the reference video as motion guidance and restyle it into a futuristic city scene, neon reflections, smooth tracking shot.
Related API-Key pages
Keep the route flat while moving between models, prompts, and related workflows.
Gemini Omni Video FAQs
What input modes does Gemini Omni support?
Gemini Omni supports text-to-video, image-to-video, and video-to-video workflows in API-Key.
Can Gemini Omni create 4K videos?
Yes. The model page exposes 720p, 1080p, and 4K options where supported.
What is Gemini Omni best for?
It is best for multimodal video work where text, image references, or an uploaded video can guide the shot.