Google's Gemini Omni 1.1 Flash adds studio-grade video control
DeepMind rolled out Gemini Omni 1.1 Flash on Wednesday, a production-ready upgrade to its generative-video model that adds the editing controls professional creators have been asking for. It's a meaningful step toward treating AI video as something you direct rather than just generate.
The headline additions are about control. Scene extension now lets the model analyze up to ten seconds of prior context — a leap over older versions, which only referenced the final second — so it can continue footage with consistent characters and narrative, extending clips in ten-second increments up to a cumulative 40 seconds. You can also specify first and last frames and have the model generate the smooth transition between them, which unlocks continuous camera orbits, zoom transitions, and looping clips. New 360p drafts render up to 60 percent faster and at roughly a third of the cost of standard 720p, making iteration and storyboarding far cheaper. When you're ready to ship, Omni 1.1 ups the output to 1080p and 4K. The API also accepts up to three seconds of video as multimodal reference input, so creators can keep visual and character consistency across takes.
Google says the model is already in production through early adopters. Adobe integrated it into Firefly for video editing, Figma Weave surfaces it for canvas-based workflows, and Runway supports it for prompt-to-video generation — with partners all pointing to that added control and reliability as the differentiator. It's available now via Google AI Studio and the Gemini Enterprise Agent Platform.
The reason this matters: AI video has spent the past couple of years getting good at one-shot generation — a solid clip from a prompt — but the bottleneck was always post-production control. Omni 1.1's direction features (frame matching, scene extension, multi-second reference context, 4K output) attack exactly that bottleneck, letting a model behave more like an editable asset in a real editing pipeline than a one-off toy. Combined with the cheaper 360p drafting tier, Google is betting that lowering the cost of exploring options wins the professional editing floor, not just the novelty crowd. That puts it in a direct contest with the video-native tools like Runway, Pika, and Luma as the market shifts from "make me a video" to "finish this shot."
What to watch: whether the multi-second memory and 4K push land in the consumer Gemini app next, and how the video-native challengers respond on price.
How much would the 40-second scene extension change your typical AI video workflow?
