On August 27, 2026, Google DeepMind released Gemini Omni 1.1 Flash, a production-ready update that adds a new suite of creative controls for generative video. The model is available now through the Gemini API in Google AI Studio and the Gemini Enterprise Agent Platform.
The update introduces four headline capabilities: scene extension up to 40 seconds with improved visual consistency, first-and-last-frame interpolation for smooth camera transitions, 360p draft mode for faster and cheaper prototyping, and upscaling to 1080p or 4K for professional output. Developers can also reference up to three seconds of existing video to maintain character and scene consistency across edits.
What's new
- Scene extension: Extend videos in 10-second increments up to a cumulative 40 seconds. The model analyzes up to 10 seconds of prior context — a leap from previous models that only referenced the final second — improving narrative flow and visual coherence.
- Keyframe control: Specify starting and ending frames to generate continuous video between them, enabling complex camera orbits, zoom transitions, and seamless loops.
- 360p draft resolution: Generate lightweight previews up to 60% faster and at roughly one-third the cost of the standard 720p resolution, suited for rapid iteration and storyboard work.
- 4K upscaling: Produce polished 1080p or 4K outputs ready for professional production pipelines.
- Video references: Include up to three seconds of reference video in multimodal prompts to preserve visual context and character consistency.
Why it matters
Gemini Omni 1.1 Flash moves generative video closer to a professional creative tool rather than a demo-grade generator. Scene extension with 10-second context windows addresses a persistent coherence problem in longer clips, while keyframe interpolation gives developers a language-native way to direct camera motion — traditionally a weak spot for diffusion-based video models. The 360p draft tier changes the economics of iteration: teams can explore multiple variations cheaply before committing to a 4K render. Combined with the Interactions API's conversational editing, the update positions Omni as a controllable layer for custom video workflows, creative tools, and media editing software.
Our take
The 10-second context window for scene extension is the most consequential technical change here; it directly tackles the temporal drift that has limited generative video to short bursts. If the consistency holds across diverse prompts, Omni 1.1 becomes viable for narrative workflows, not just isolated clips. The pricing table (Omni 1.1 Flash highlighted against Veo 3.1 variants) suggests Google is positioning this as the quality-per-dollar option for controllable video, though real-world cost at scale will depend on the mix of draft vs. final renders.
Series: 1. Koray Kavukcuoglu Named DeepMind SVP as Hassabis Becomes Chair · 2. Google DeepMind launches Gemini Omni 1.1 Flash with scene extension, keyframe control, and 4K upscaling · DeepMind Leadership
