Google has released Gemini Omni 1.1 Flash, a production-ready update for developers. To begin with, it gives you more control over generative video. For example, you can extend scenes, set start and end frames, and create sharp 4K output.
Furthermore, the model is ready for real-world use through the Gemini API in Google AI Studio. Whether you build video workflows, creative tools, or media editing software, these updates help you work faster and deliver polished results. So, here’s what’s new.
Extend Scenes for Longer Storytelling
To start, scene extension lets you take an existing video and keep generating footage from where it stopped. As a result, the clip continues smoothly, so your story flows.
In addition, Omni 1.1 now reads up to 10 seconds of earlier context. In contrast, previous models only looked at the final second. Because of this bigger view, your visuals stay consistent and your story stays on track. Meanwhile, you can extend videos in 10-second steps, up to 40 seconds total. Ultimately, you can use this to build longer stories or branch into fresh creative directions.
Specify First and Last Frames
Next, you can set the starting and ending frames of a shot. Then, Omni 1.1 fills in continuous video between the two keyframes.
As a result, you get smooth transitions and controlled camera movements. For instance, it works well for complex camera orbits, zoom transitions, and seamless looping clips. In short, you decide where the shot begins and ends, and the model handles the rest.
Draft Videos Faster in 360p
Of course, speed matters when you test ideas. For that reason, Omni 1.1 lets you generate lightweight previews in 360p resolution. Better yet, these drafts render up to 60% faster and cost a third of what 720p costs.
Therefore, use this mode for rapid prototyping, storyboard iteration, and quick rendering. First, try several versions. Then, change one thing at a time, and compare them side by side. Finally, once you like a draft, move it forward.
Upscale to 4K Resolution
Later, when your project is ready, you can polish it. Specifically, Omni 1.1 produces clean 1080p or 4K outputs built for professional production. As a result, your final footage looks sharp and ready to ship.
Add Video References in Your Input
In addition, you can reference up to three seconds of video when you craft a scene. Consequently, this helps you keep visual context and character consistency across your work.
For example, feed the model reference clips, and it holds onto those details. As a result, your characters stay the same, and your scenes match the look you want.
Ideas for What You Can Build
Naturally, these new tools open up practical projects. So, here are a few ways developers put them to work.
- Transition tools: First, drop in a first and last frame. Then, generate the transition with presets or a prompt box. As a result, Omni’s full-context reasoning gives you camera moves that feel real.
- Property walkthroughs: Meanwhile, you can move the camera through rooms. For instance, it arcs, pushes in, and pulls back. In addition, it shows a home at its best, without adding furniture or details that don’t exist.
- Draft rooms: On the other hand, creators often make several videos before they find the right one. So, generate three or four drafts in 360p, tweak one variable at a time, and compare the results cheaply.
How Customers Use Omni Flash
Already, teams run Gemini Omni Flash in production through the Agent Platform API. Here’s what they say.
To begin with, Adobe built Gemini Omni Flash into Adobe Firefly to power its video editing features.
Likewise, Figma Weave uses it to help creative teams direct their work. “Gemini Omni Flash is one of the strongest video models available in Figma Weave, where the canvas helps creative teams build on every generation,” said Itay Schiff, Creative Director at Figma Weave. “With extensions, richer reference material, and 4K resolution, Gemini Omni Flash takes teams beyond generating videos to truly directing them.”
In addition, GMI Cloud values the model’s accuracy. “What stands out about Gemini Omni Flash is its accuracy: the details hold up under scrutiny,” said Louisa Guo, VP of Marketing at GMI Cloud. “For customers creating educational and explanatory content, where getting things right is essential, that reliability matters more than any single feature.”
Finally, Runway fits it into existing workflows. “Omni Flash fits naturally into how people already use Runway: start with a prompt, an image or a video, then generate or edit from there,” said Jamie Umpherson, Chief Creative Officer at Runway.
Start Building Today
Ultimately, Omni 1.1 is rolling out across the Google developer ecosystem. So, here’s how to get started.
- Google AI Studio: First, try Omni 1.1 directly in Google AI Studio.
- Gemini Enterprise Agent Platform: Meanwhile, enterprises can build with Omni 1.1 through the Agent Platform API.
- Developer documentation: In addition, check the official docs, the cookbook, and the prompting guides. They show you how to add scene extensions, video references, and upscaling to your apps.
Moreover, Omni 1.1 is available to all Google AI Plus, Pro, and Ultra subscribers worldwide in Google Flow, starting today. Likewise, scene extension is available to those same subscribers in the Gemini app.
