Gemini Omni 1.1 Flash lets you build with more control

2026-08-28 · Google DeepMind

Release overview

Google DeepMind has launched Gemini Omni 1.1 Flash, a production-ready update for developers building with generative video. The focus is clear: better control, faster iteration, and output quality suitable for real deployment.

The model is available through the Gemini API in Google AI Studio and the Gemini Enterprise Agent Platform, targeting use cases such as generative video workflows, creative tools, and media editing software.

What’s new in Omni 1.1 Flash

1) Scene extension for longer storytelling

Omni 1.1 Flash can continue an existing video seamlessly:

  • Extension in 10-second increments
  • Up to 40 seconds total cumulative length
  • Uses up to 10 seconds of prior context (up from earlier behavior that referenced only the final second)

This improves visual consistency and narrative adherence when extending clips, making longer story arcs and branching creative directions more practical.

2) First-and-last-frame control

Developers can now specify starting and ending keyframes for a shot, and Omni 1.1 Flash generates continuous motion between them. This is useful for:

  • Complex camera moves (orbits, zoom transitions, dolly-style effects)
  • Smooth transitions
  • Continuous-shot style sequences
  • Seamless looping clips

In short, the update gives creators stronger shot-level control rather than relying only on text prompts.

3) Faster, cheaper 360p drafts

To speed up experimentation, the model supports 360p preview generation:

  • Up to 60% faster than standard 720p rendering (as stated by Google)
  • Around one-third of the cost of 720p

This enables a practical workflow: test ideas quickly at low resolution, then move to higher-quality output once direction is validated.

4) 4K upscaling for final delivery

After iteration, projects can be upscaled to crisp 4K, aiming at polished, professional-looking final results.

Why this matters

Google positions Omni 1.1 Flash as a step from “can generate” to “ready for production.” The update combines creative flexibility with operational efficiency:

  • Longer coherent scene continuation
  • Explicit control over shot boundaries
  • Lower-cost rapid prototyping
  • High-resolution finishing for delivery

For teams building AI-native video products, these additions make generative video more controllable and deployable in real-world pipelines.

Source