saved
Gemini Omni 1.1 Flash lets you build with more control
gist
Google DeepMind product managers Anish Nangia and Alisa Fortin ship Gemini Omni 1.1 Flash as a production-ready generative video model on the Gemini API. Scene extension now reads 10 seconds of prior context and can grow a clip in 10-second steps to 40 seconds. First-and-last-frame interpolation, 360p drafts at a third of 720p cost, 1080p or 4K upscaling, and three-second video references are the new directing controls. Output costs $0.03 per second at 360p and $0.30 at 4K. Adobe Firefly, Figma Weave, GMI Cloud, and Runway already ship Omni Flash.
ideas
- Production-ready on the Gemini API. Omni 1.1 targets generative video workflows and media editors in Google AI Studio, Gemini Enterprise Agent Platform, Flow, and the Gemini app.
- Scene extension keeps 10 seconds of context. Prior models used only the last second; 1.1 continues a clip in 10-second increments up to 40 seconds.
- Direct with keyframes and references. First and last frames interpolate orbits and loops; up to three seconds of reference video hold character and motion consistency.
- Draft cheap, finish in 4K. 360p previews are up to 60% faster and a third the 720p price, $0.03 per second versus $0.10; 1080p is $0.15 and 4K is $0.30.
- Customers treat it as a director. Adobe Firefly, Figma Weave, GMI Cloud, and Runway report using Omni Flash for editing, branching, and educational accuracy.
quotes
“today’s updates make Omni 1.1 production-ready for professional use via the Gemini API in Google AI Studio.”
“the model can now analyze up to 10 seconds of prior context — a leap from previous models that only referenced the final second.”
“You can extend videos in 10-second increments up to a total cumulative length of 40 seconds.”
“takes teams beyond generating videos to truly directing them.”