Gemini Omni Flash Reaches General Availability
Gemini CLIGoogle has moved Gemini Omni Flash out of preview and into general availability, shipping gemini-omni-1.1-flash as the production-ready version of its fast, conversational video generation and editing model. The GA release adds video extension for continuing existing clips, image-to-video interpolation between two frames, and a configurable resolution parameter supporting up to 4K output. The preview endpoint gemini-omni-flash-preview will be deprecated on September 30, 2026, giving developers about a month to migrate.
Key Takeaways
- Gemini Omni Flash graduates from preview to GA, shipping as
gemini-omni-1.1-flashfor production use on the Gemini API after roughly two months in public preview. - Video extension now analyzes up to ten seconds of prior footage (versus just the last second previously), letting developers extend clips in 10-second increments up to 40 seconds total.
- Image-to-video interpolation via the
image_to_videotask lets developers set start-and-end frames and have the model generate a smooth transition between them, including complex camera moves. - A new resolution parameter in
video_configsupports 360p, 720p (default), 1080p and 4K output, alongside a faster, cheaper 360p draft mode for quick previews. - The preview endpoint
gemini-omni-flash-previewis being deprecated on September 30, 2026, putting a firm deadline on migrating existing integrations to the GA model. - The release drew organic coverage on Hacker News and Techmeme shortly after launch, with observers noting the GA ship followed an extended pre-GA testing period.
Sources & Mentions
5 external resources covering this update
Gemini Omni 1.1 Flash
Hacker News
Google launches Gemini Omni 1.1 Flash, which it says delivers studio-quality video production, including the ability to extend a scene, 4K upscaling, and more
Techmeme
Gemini Omni can now extend videos to 40 seconds and upscale them to 4K
Digital Trends
Google's Gemini Omni 1.1 Flash makes AI video generation cheaper and more flexible
The Decoder
GOOGLE: Gemini Omni 1.1 Flash has been officially announced and now available on Gemini APIs and Google AI Studio (@testingcatalog)
X (Twitter)
Gemini Omni Flash Graduates to General Availability
Google has officially released gemini-omni-1.1-flash, the general-availability version of Gemini Omni Flash, its fast, conversational video generation and editing model. The model had been in public preview since June 2026, and the GA release brings a substantial set of new creative controls aimed squarely at developers building automated video and media-editing workflows on the Gemini API.
New Video Extension and Interpolation Capabilities
The headline addition is video extension, accessed through a new extend task. Rather than only looking at the final second of a clip, Gemini Omni Flash now analyzes up to ten seconds of prior footage before generating a continuation, which Google says produces noticeably more consistent motion, lighting, and narrative continuity. Clips can be extended in ten-second increments up to a total of forty seconds, making it practical to grow a short generated clip into something closer to a finished scene.
Alongside extension, the GA release introduces interpolation between two images via the image_to_video task. Developers can now supply up to two images and have the model generate a video that transitions naturally between them — useful for specifying a precise start and end frame and letting the model fill in a smooth, continuous shot, including complex camera moves like orbits.
Configurable Resolution Output
A new resolution parameter inside video_config gives developers direct control over output quality, supporting 360p, 720p (the default), 1080p, and 4k. Google also added a faster and cheaper 360p draft mode, intended for quickly previewing a shot before committing to a full-resolution render — reported to run up to 60% faster at roughly a third of the cost of standard output.
Migration and Deprecation
With the GA launch, the earlier preview endpoint gemini-omni-flash-preview is now on a deprecation clock: Google says it will be shut down on September 30, 2026, and developers still calling the preview endpoint should migrate to gemini-omni-1.1-flash before then. The model is accessible through the Gemini API, Google AI Studio, and the Gemini Enterprise Agent Platform.