What Is Gemini Omni 1.1 Flash? Google's Video Model
The Short Answer
Gemini Omni 1.1 Flash is Google’s video generation and editing model, announced August 27, 2026 and callable as gemini-omni-1.1-flash.
It is a point release, not a new architecture — but the point release fixes the thing that made the previous version awkward to build on. Scene extension now reads up to ten seconds of the existing clip instead of just the last second, so extended footage stays consistent instead of drifting away from the characters and motion you started with.
The commercially interesting part is the price ladder: $0.03 per second at 360p for drafts, rising to $0.30 per second at 4K for finals.
What Actually Changed in 1.1
| Capability | Gemini Omni Flash | Gemini Omni 1.1 Flash |
|---|---|---|
| Base clip length | up to 10s | up to 10s |
| Scene extension | analyses last ~1s | analyses up to 10s |
| Max extended length | — | 40s (10s increments) |
| Style reference | — | up to 3s of external footage |
| Keyframe control | — | start + end frame |
| Draft tier | — | 360p, ~60% faster |
| Resolutions | 720p | 360p / 720p / 1080p / 4K |
Four things are genuinely new for developers:
- Longer-context scene extension. The model looks back across up to ten seconds of generated footage when continuing a shot. This is the difference between a 40-second sequence and four unrelated 10-second clips stitched together.
- External footage as style reference. You can upload up to three seconds of your own video to carry over a character’s appearance or a motion pattern into generated shots.
- Start and end frame locking. Supply two keyframes and the model generates the camera movement between them — a deterministic handle on a notoriously non-deterministic medium.
- A draft tier that is actually cheap. 360p runs up to 60% faster at about one third the cost of 720p. You iterate on composition at draft price and pay full rate once.
The Pricing Ladder
Per output second, as announced August 27, 2026:
| Resolution | Price / second | 10s clip | 40s sequence |
|---|---|---|---|
| 360p (draft) | $0.03 | $0.30 | $1.20 |
| 720p | $0.10 | $1.00 | $4.00 |
| 1080p | $0.15 | $1.50 | $6.00 |
| 4K | $0.30 | $3.00 | $12.00 |
Note that 1080p and 4K are upscaled outputs, not natively rendered at that resolution. That matters for expectations: 4K here buys you resolution and deliverable format, not four times the detail the model actually synthesised.
How It Compares Inside Google’s Own Lineup
Google now sells video at several price points, and Omni 1.1 Flash sits deliberately between them:
| Model | 720p | 1080p | 4K |
|---|---|---|---|
| Gemini Omni 1.1 Flash | $0.10 | $0.15 | $0.30 |
| Veo 3.1 Lite | $0.05 | $0.08 | — |
| Veo 3.1 Fast | $0.10 | $0.12 | $0.30 |
| Veo 3.1 Standard | ~$0.40 | ~$0.40 | — |
Veo 3.1 Lite is still cheaper per second. Omni 1.1 Flash is not competing on raw price against Google’s own budget tier — it is competing on editing control. Extension, keyframes and style reference are the reason to pay the difference. If you need one 8-second shot and nothing else, Lite is the rational call.
Where You Can Use It
- Google AI Studio — interactive, fastest way to evaluate output quality
- Gemini API —
gemini-omni-1.1-flash, documented under the Gemini API omni docs - Gemini app and Google Flow — consumer-facing surfaces for non-developers
There is no local deployment path. No open weights, no Hugging Face release, no self-hosting. If your constraint is data residency or air-gapped operation, this model is not an option regardless of price.
Who Should Care
Use it if you are building a video pipeline where a shot gets revised repeatedly — marketing variants, explainer sequences, ad iterations. The draft tier plus extension is a genuine workflow, not a spec-sheet feature.
Skip it if you need single one-off clips at the lowest possible price (Veo 3.1 Lite, or Alibaba’s Wan 3.0 at $0.20 per 1080p second with native 30-second clips), or if you need weights you can run yourself.