- Text, up to 3 images, short video
Inputs
- 10 seconds
Clip length
- 720p
Resolution
- 16:9 or 9:16
Aspect ratios
About Gemini Omni Flash
Gemini Omni Flash is Google's video model for generation and conversational editing, released in public preview as gemini-omni-flash-preview. It is a video model, not an image model. In Masonry it runs on Google's Vertex AI and covers three jobs: text-to-video from a prompt, image-to-video from up to three input images, and editing a short source video according to a prompt. Google says it generates video with audio, and that Omni currently produces 10-second generations. In Masonry the clip is a fixed 10 seconds at 720p, in 16:9 or 9:16.
The controls are deliberately simple. There is no duration or resolution setting, so every clip is the same length and size, which keeps drafts predictable. That makes Omni Flash useful for quick video ideas, vertical social drafts, and short edits to an existing clip. When a brief needs a longer clip, a fixed end frame, or 1080p, models such as WAN 3.0, FLUX 3 Max, or Kling O3 offer those controls. Every Omni Flash video carries Google's SynthID watermark. Gemini Omni Flash 1.1, the GA version, is the newer model with more controls.
Why teams choose Gemini Omni Flash
Choose Gemini Omni Flash when you want a quick, sound-on 10-second clip from text, images, or a short source video, without tuning duration or resolution. Its editing mode sets it apart from text-and-image-only video models. For longer clips, 1080p output, or a fixed end frame, WAN 3.0, FLUX 3 Max, and Kling O3 give you more control. For most new work, use Gemini Omni Flash 1.1, Google's GA version, which adds 3 to 10 second durations, 360p to 4K output, and an optional end frame. Keep this preview model for its short source-video editing.
What Gemini Omni Flash can do
The capabilities that set Gemini Omni Flash apart and earn its place in a brief
Text, Image, or Video Input
Generate from a text prompt, from up to three input images, or from a short source video that Omni Flash edits according to your prompt instead of starting from scratch.
Video Editing by Prompt
Supply a source video of up to 3 seconds in Masonry and describe the change. Google describes Omni as built for conversational video editing in natural language.
Native Audio
Google says the model generates video with audio based on your description, so sound arrives with the picture.
Fixed 10-Second Clips
Every clip is 10 seconds, the length Google says Omni currently generates. There is no duration setting to choose.
720p in 16:9 or 9:16
Masonry delivers 720p output in landscape 16:9, the default, or vertical 9:16 for Reels, TikTok, and Shorts.
SynthID Watermarking
Google applies SynthID watermarking to all generated videos.
Where teams reach for Gemini Omni Flash
- Quick text-to-video drafts for social and ad ideas
- Vertical 9:16 clips for Reels, TikTok, and YouTube Shorts
- Animating product or campaign images into short clips
- Prompt-based edits to a short existing clip
- Sound-on concept videos for pitches and storyboards
Frequently asked questions
What teams need to know about creating with Gemini Omni Flash in Masonry
What is Gemini Omni Flash?
Gemini Omni Flash is Google's video model for generation and conversational editing, available in public preview as gemini-omni-flash-preview. In Masonry it runs on Google's Vertex AI and makes video from text, images, or a short source video.
Is Gemini Omni Flash an image or a video model?
It is a video model. It accepts images as inputs, but its output is always a video clip.
How long is a Gemini Omni Flash video?
10 seconds. Google says Omni currently generates 10-second videos, and Masonry doesn't offer a duration setting.
What resolution and aspect ratios does Gemini Omni Flash support?
Masonry delivers 720p in 16:9, the default, or 9:16.
Can Gemini Omni Flash edit an existing video?
Yes. Supply a source video of up to 3 seconds in Masonry and describe the change, and Omni Flash edits it instead of generating from scratch.
Does Gemini Omni Flash generate audio?
Yes. Google says the model generates video with audio based on your description.
Can Gemini Omni Flash animate an image?
Yes. Add up to three input images with your prompt and describe the motion you want.
Are Gemini Omni Flash videos watermarked?
Yes. Google applies SynthID watermarking to all generated videos.
How much does Gemini Omni Flash cost in Masonry?
Masonry bills each clip in credits, and the editor shows the exact cost before you generate.
How is Gemini Omni Flash different from Gemini Omni Flash 1.1?
Both are available in Masonry. Gemini Omni Flash 1.1 is the newer version with more output options. This preview model gives fixed 10-second 720p clips and adds editing of a short source video.
How does my team use Gemini Omni Flash in Masonry?
Open a Masonry canvas, pick Gemini Omni Flash from the model selector, and describe the video you need: a product shot, an ad creative, a social post. Masonry generates it, then you refine, edit, and combine Gemini Omni Flash with other models in one workspace.
Is Gemini Omni Flash free to try?
New Masonry accounts get 50 free credits to try GPT Image 2.5 Flare. Gemini Omni Flash is included in every paid plan, and the editor shows the exact credit cost before you generate.
How do I write good prompts for Gemini Omni Flash?
Describe one clear scene that fits 10 seconds, naming the subject, action, camera move, setting, and the sound you want. For image-to-video, describe the motion rather than re-describing the image. For an edit, describe only the change you want made to the source video.
Who makes Gemini Omni Flash?
Gemini Omni Flash is built by Google. Inside Masonry it runs alongside 50+ image and video models, so your team can pick the right one for each brief without switching tools.
Start creating with Gemini Omni Flash
Generate, edit, and compare across 50+ models in one workspace.
Guides for Gemini Omni Flash
Prompt walkthroughs and examples from the Masonry blog
Explore more AI models
Compare Gemini Omni Flash with other models teams run in Masonry