Gemini Omni Flash is Google's fast model for making and editing short videos with sound from text, images and video in the Gemini API.
In plain words
The current version, Gemini Omni 1.2 Flash, makes clips of up to 15 seconds with matching audio. You can change a clip over several turns in plain language instead of starting again. Up to six reference pictures keep a character or product looking the same from shot to shot. Google now points developers who want video in the Gemini API to this model, and released version 1.2 on 5 November 2026 at a 30% lower price than 1.1.
A quick example
Google's own docs show a drawing of a fish leaping from water used as a guide. The prompt asks for realistic footage that follows the drawing's movement without showing the drawing itself.
Where you'll meet it
- Short videos, product clips and ads made through the Gemini API.
- Editing a generated clip by asking for changes in conversation.
Related
Sources
- DocsGenerate and edit videos with Gemini Omni Flash, Google · read 5 Nov 2026
- DocsGemini Omni 1.2 Flash: Fast and cost-effective with longer clips, Google DeepMind · read 5 Nov 2026
- DocsVideo generation in the Gemini API, Google · read 5 Nov 2026
Checked by WhatIsAI against the sources above · 5 Nov 2026. Spotted an error? Report it.