Google DeepMind ships Gemini Omni 1.1 Flash for video developers

Google DeepMind

Video official + media 2 src. ~1 min

An iteration of Gemini’s video model focused on developer controls: scene extension that pulls up to 10 seconds of prior context and extends in 10-second increments to 40 seconds, first-and-last-frame generation for orbits and seamless loops, 360p drafts that preview up to 60% faster and at one-third the cost of 720p, 4K upscale on top of the 1080p output, and up to 3 seconds of video reference for character and visual continuity.

Why it matters

Brings explicit long-form scene stitching and 4K finish to Gemini video at a 360p draft price point, narrowing the gap with Sora-class long-clip tooling for developers building through Google AI Studio, the Gemini Enterprise Agent Platform API, and Google Flow.

Importance: 2/5

official confirmation

text-to-video image-to-video video-to-video scene-extension first-last-frame 4k-upscaling

Sources

official Build with Gemini Omni 1.1 Flash

secondary Gemini Omni 1.1 Flash model card notes (third-party picker)