Text to video
Describe a shot in plain language and get a moving scene back, with camera direction, lighting and pacing taken from the prompt.
gemini-omni-1.1-flash
Turn a prompt or a picture into directable video
Gemini Omni 1.1 Flash generates video from text, images and reference media, then keeps editing the result in a conversation. Write a prompt below to open the workspace.
This page is an independent product walkthrough. The prompt box is a preview and does not render video by itself.
Gemini Omni 1.1 Flash video examples
Short clips generated from text, still images, reference media and first-to-last-frame interpolation. Every example below started as a one-line brief.
What Gemini Omni 1.1 Flash can do
One model for text, image and reference-driven video generation, with an editing loop that keeps the same clip instead of restarting from scratch.
Image to video
Animate a still: keep its composition and subject, then add the motion, atmosphere and camera move you ask for.
Reference media
Pass reference images or video clips so characters, products and art direction stay consistent across shots.
First and last frame
Supply a starting frame, an ending frame, or both, and the model interpolates the motion between them.
Conversational editing
Keep editing through the Interactions API: ask for warmer light, a tighter frame or a different ending and the model revises the clip.
Extend and upscale
Generate 3-10 second clips, continue a scene past its original ending, and upscale drafts when you are happy with the cut.
From a prompt to a finished cut
Four steps that match how the model is used in Google AI Studio, the Gemini API and Gemini Enterprise Agent Platform.
Start a draft
Write a prompt, or upload an image, a reference clip or a first and last frame to anchor the shot.
Pick the format
Choose 9:16 or 16:9 and render a 360p draft, then move to 720p, 1080p or 4K when the edit is locked.
Direct in conversation
Ask for changes through the Interactions API: lighting, pacing, camera movement, ending, wardrobe.
Extend and export
Continue the scene, stitch the best takes into a sequence, and download the final render.
Pick the entry point that matches your material
Gemini Omni 1.1 Flash accepts four kinds of input. Choosing the right one up front saves the most iterations.
InputModeWhen to use it
Prompt-onlyText to videoFastest path: a short brief becomes a clip with no source material.
Still imageImage to videoBest when the composition or product shot already exists and only motion is missing.
Reference setSubject consistencyUse reference images or clips when a character or product must survive several shots.
Frame controlFirst and last frameUse two frames when the shot has to land on an exact pose, product angle or title card.
ConversationIterative editingUse the Interactions API when the clip exists and you want to revise instead of regenerate.
Where teams use Gemini Omni video
From a single vertical ad to a previsualized scene, the same model covers drafts, revisions and finals.
Vertical hooks and cutdowns
Render 9:16 hooks in a draft resolution, then upscale the winners for paid placement.
Previsualization
Block a scene with first and last frames before a shoot day to test timing and camera moves.
Ecommerce motion
Turn catalogue stills into looping product motion while the label and packaging stay accurate.
Animatics
Extend a storyboard panel into a few seconds of motion to pitch pacing before animation.
Campaign variants
Keep one look and produce alternate endings, hooks and framing for a launch.
Client revisions
Apply feedback as a conversation instead of rebuilding the shot from the prompt.
Choose a Gemini Omni plan
Monthly and annual subscriptions, plus a one-time credit booster when a launch needs extra renders. An active subscription is required to generate.
Starter
$9 per month
300 credits / month
- Text and image to video
- 720p, 9:16 and 16:9
- 3-10 second clips
Pro
$19 per month
1,000 credits / month
- Everything in Starter
- First and last frame interpolation
- Extend scenes up to 40s
- 1080p upscaling
Premium
$49 per month
3,000 credits / month
- Everything in Pro
- 4K upscaling
- 360p fast drafts
- Priority render queue
Annual billing is available at $90 / $190 / $490. Need extra renders mid-cycle? A $29 booster adds 300 credits for 90 days.
Gemini Omni 1.1 Flash questions
Quick answers on inputs, editing, output formats, access paths and regional limits.
What is Gemini Omni 1.1 Flash?+
Gemini Omni 1.1 Flash (gemini-omni-1.1-flash) is a video generation model that turns text, images and reference media into short video, and then keeps editing that video in a conversation.
Can it generate video from text alone?+
Yes. A written prompt can describe the subject, the action, the camera movement and the lighting, and the model renders the scene without any source media.
Can it turn an image into video?+
Yes. Upload a still and describe the motion you want. The model keeps the composition and subject of the image and animates it.
What are reference images and reference video used for?+
Reference media anchors things that must stay consistent between shots, such as a character design, a product label or an art direction.
How does first and last frame interpolation work?+
You can provide a starting frame, an ending frame, or both. The model fills in the motion between them, which is useful when a shot has to land on an exact pose or product angle.
What is conversational editing?+
Through the Interactions API you keep working on the clip you already have. Asking for warmer light, a tighter frame or a different ending revises the video instead of starting over.
How long can a clip be?+
A single generation produces a 3-10 second clip. A scene can be continued past its original ending, and extensions can take a sequence up to about 40 seconds.
Which resolutions are supported?+
360p drafts render quickly for iteration, 720p is the standard output, and finished shots can be upscaled to 1080p or 4K.
Which aspect ratios can I use?+
9:16 for vertical and 16:9 for widescreen are both supported, so the same shot can serve social placement and long-form edits.
Is there a file size limit for uploads?+
Inline uploads are limited to a few megabytes; larger files must be passed as a URI rather than inline data.
Where can I access Gemini Omni 1.1 Flash?+
The model is available through Google AI Studio, the Gemini API and the Gemini Enterprise Agent Platform, depending on whether you are prototyping or shipping production volume.
Are there regional restrictions?+
Editing uploaded video is restricted in the EEA, the UK and Switzerland, so those accounts cannot use the video editing paths on uploaded material.
Do I need a subscription to use this site?+
Yes. An active subscription is required, and accounts without one are sent to the pricing page. Sign-in is Google only.
Is geminiomni.lat an official Google site?+
No. geminiomni.lat is an independent third-party website and is not affiliated with, endorsed by, sponsored by or operated by Google.
Turn your idea into video
Start with a prompt, a still or a pair of frames. Subscribe, generate a draft, and direct the shot from there.