Google Releases Gemini Omni 1.1 Flash with New Video Controls

Google expands Gemini Omni for controllable generative video

Google introduced Gemini Omni 1.1 Flash on August 27, 2026, adding developer-focused controls for generative video.

Google's launch post describes Omni 1.1 as production-ready for professional workflows through the Gemini API in Google AI Studio. At the same time, Google Cloud's Gemini Enterprise Agent Platform documentation lists the enterprise model identifier gemini-omni-1.1-flash-preview with a Preview launch stage. Developers should therefore check the terms and model identifier for the specific Google surface they use rather than assuming every endpoint has the same availability status.

What is new in Omni 1.1 Flash

The update adds several controls aimed at editing and longer-form video workflows:

• Scene extension: the model can analyze up to 10 seconds of prior context, compared with the final second in the earlier system, and extend a video in 10-second increments up to 40 seconds of cumulative length.
• First- and last-frame interpolation: developers can provide two keyframes and ask Omni 1.1 to generate the continuous transition between them.
• 360p drafts: Google says 360p previews can generate up to 60% faster than 720p based on system throughput and at one-third of the 720p cost.
• Higher-resolution output: Omni 1.1 supports 1080p and 4K output.
• Video references: a prompt can reference up to three seconds of video to help preserve visual context and character consistency.

These capabilities target creative tooling, video editing, storyboarding and media-production applications rather than only one-shot text-to-video generation.

Availability requires a little nuance

Google says Omni 1.1 is rolling out through Google AI Studio and the Gemini Enterprise Agent Platform. It is also available in Google Flow for Google AI Plus, Pro and Ultra subscribers, with scene extension available in the Gemini app for those tiers.

For the enterprise platform, Google's documentation currently lists Gemini Omni 1.1 Flash (Preview) and the model ID gemini-omni-1.1-flash-preview. The documentation says the preview supports video, image and text tasks, with supported video resolutions of 360p, 720p, 1080p and 4K.

That distinction matters: the launch article uses 'production-ready' to describe the update, while a specific Google Cloud deployment surface remains governed by Preview terms.

Why this release matters

AI video systems are increasingly being used inside editing and creative applications, where repeatability and control matter as much as raw visual quality. Scene continuation, explicit endpoint frames and low-cost draft renders can reduce the number of full-quality generations required to reach a usable result.

Google also highlights real production integrations around Gemini Omni Flash, including Adobe Firefly and other creative platforms. Those customer examples show the broader direction: video models are becoming programmable media components rather than isolated generation demos.

For developers evaluating Omni 1.1, the most important questions are which API surface they use, the applicable preview or production terms, supported resolutions, generation cost and whether the new continuity controls fit their editing workflow.