Mixer AI
Gemini Omni
Create a short scene or transform footage you already have, choosing the version and input that fit.
When to choose it
Gemini Omni is not only for new video. Take an existing moment and ask for a different setting or visual treatment. For a new scene, use text, image references or, in version 1.1, first and last frames. Choose the workflow around your source material, not just the version number.
- New generation and video transformation share one family.
- Version 1.1 adds a dedicated frame mode and 360p drafts.
Modes and settings
- 1.0 and 1.1 Flash support text, up to 7 photo references and video transformation. Version 1.1 also supports first and last frames.
- New clips: 4, 6, 8 or 10 seconds, at 720p, 1080p or 4K. Version 1.1 also offers 360p.
- Video transformation operates on a segment up to 10 seconds and has a separate pricing method.
Price in coins
| Version | Quality / size | Input | Duration | Coins / generation |
|---|---|---|---|---|
| 1.1 Flash | 360p | Video transformation | Segment up to 10s | 102 |
| 1.1 Flash | 360p | Text or images | 4 s | 40 |
| 1.1 Flash | 360p | Text or images | 6 s | 52 |
| 1.1 Flash | 360p | Text or images | 8 s | 65 |
| 1.1 Flash | 360p | Text or images | 10 s | 77 |
| 1.1 Flash | 720p | Video transformation | Segment up to 10s | 102 |
| 1.1 Flash | 720p | Text or images | 4 s | 40 |
| 1.1 Flash | 720p | Text or images | 6 s | 52 |
| 1.1 Flash | 720p | Text or images | 8 s | 65 |
| 1.1 Flash | 720p | Text or images | 10 s | 77 |
| 1.1 Flash | 1080p | Video transformation | Segment up to 10s | 102 |
| 1.1 Flash | 1080p | Text or images | 4 s | 40 |
| 1.1 Flash | 1080p | Text or images | 6 s | 52 |
| 1.1 Flash | 1080p | Text or images | 8 s | 65 |
| 1.1 Flash | 1080p | Text or images | 10 s | 77 |
| 1.1 Flash | 4K | Video transformation | Segment up to 10s | 152 |
| 1.1 Flash | 4K | Text or images | 4 s | 90 |
| 1.1 Flash | 4K | Text or images | 6 s | 102 |
| 1.1 Flash | 4K | Text or images | 8 s | 114 |
| 1.1 Flash | 4K | Text or images | 10 s | 127 |
| 1.0 Flash | 720p | Video transformation | Segment up to 10s | 102 |
| 1.0 Flash | 720p | Text or images | 4 s | 40 |
| 1.0 Flash | 720p | Text or images | 6 s | 52 |
| 1.0 Flash | 720p | Text or images | 8 s | 65 |
| 1.0 Flash | 720p | Text or images | 10 s | 77 |
| 1.0 Flash | 1080p | Video transformation | Segment up to 10s | 102 |
| 1.0 Flash | 1080p | Text or images | 4 s | 40 |
| 1.0 Flash | 1080p | Text or images | 6 s | 52 |
| 1.0 Flash | 1080p | Text or images | 8 s | 65 |
| 1.0 Flash | 1080p | Text or images | 10 s | 77 |
| 1.0 Flash | 4K | Video transformation | Segment up to 10s | 152 |
| 1.0 Flash | 4K | Text or images | 4 s | 90 |
| 1.0 Flash | 4K | Text or images | 6 s | 102 |
| 1.0 Flash | 4K | Text or images | 8 s | 114 |
| 1.0 Flash | 4K | Text or images | 10 s | 127 |
Coins are Mixer AI balance units. The table shows a base request without extra variations. Multiple results and new attempts can increase the total; check it before submitting.
Video transformation has a flat charge for a segment of up to 10 seconds. Other rows create a new clip of the selected length. These are different billing methods.
What you can make
- Change the time of day in a short filmed scene.
- Connect two compositions prepared in advance.
How to direct the result
- For edits, begin with what must stay: character, action and camera.
- Do not mix the first/last-frame workflow with an ordinary reference set.
Example: evening instead of daylight
Change the uploaded scene from daylight to early evening, with warm shop-window light and a cool sky. Keep the people, clothes, movement and camera angle. Add no new objects. Window reflections should match the evening lighting.
What to keep in mind
- The duration setting for a new clip is not a duration promise for a video transformation.
- Changing versions changes available modes. Check the input files after switching.
Sources
- Prices and availability — Mixer AI catalog, updated 2026-09-18.
- Official maker site: Google
Model facts verified: 2026-09-20.
FAQ
When do I need version 1.1?
For the dedicated first/last-frame mode or 360p drafts.
Can I transform a whole one-minute video?
This workflow processes a segment up to ten seconds. Select the relevant excerpt first.
Is video editing billed per second?
This integration uses a separate flat price based on settings. Do not confuse it with new-clip pricing.
Can I add an ending frame in 1.0?
The dedicated first/last-frame mode is available in 1.1.
How should I preserve appearance?
Prepare consistent photos and explicitly identify what must remain. No mode guarantees exact identity.
Does 4K improve instruction accuracy?
Not by itself. It sets output size; check the edit or movement first.
What changes the price?
It depends on version, quality and mode. The table prices one generation at the listed settings. Additional results, new attempts and chargeable input materials can increase the total.
Do I need a separate model subscription?
No. In Mixer AI, you top up one coin balance and use it for the tasks you need. You do not need a separate subscription to this model.