Google has made Gemini Omni 1.2 Flash generally available, turning its AI video model into a production tool for the first time. The model, announced on November 5, generates and edits video clips up to 15 seconds long, up from 10 seconds, and costs 30 percent less than its predecessor at $0.07 per second of 720p video. It is available now through the Gemini API and Gemini for business, with wider rollouts across Flow, Google Vids, the Gemini app and Google Ads.
Longer clips, lower cost
The headline change is duration. Omni 1.2 Flash creates and edits clips natively up to 15 seconds, with length set in one-second steps from 3 to 15 seconds. That is long enough for a complete product explainer, a virtual try-on sequence or a full social ad spot in a single generation, without stitching shorter clips together.
The price cut matters as much as the length. At $0.07 per second of 720p video, a 15-second clip costs about a dollar, and Google says editing a detail instead of re-generating the clip keeps costs down further. For businesses making video ads, demos and try-on clips, that pricing turns a demo tool into something they can run every day.
Editing in conversation
The second change is how editing works. Omni 1.2 Flash supports multi-turn conversational editing across 15-second clips, so a creator can swap a product or background, change the lighting or weather, or switch the aspect ratio turn by turn instead of starting over. Google’s example walks one scene through five backgrounds while the characters and camera framing stay put.
Reference following gets stronger too. Creators can guide generation with character, product or style images, motion clips of up to three seconds and audio cues, and the model holds those details from start to finish. Faces, garments, packaging typography and brand colours stay consistent across camera angles, making virtual try-on clips and branded video ads practical rather than just impressive.
Speech and camera work get better too. Scenes can carry more fluid, natural-sounding dialogue with clearer lip-sync and distinct voices for different speakers, and complex camera moves like sweeping flythroughs and zoom-outs now come with fewer unwanted jump cuts. Google’s example is one continuous drone flight through the Pantheon, from the piazza to the oculus, with no cuts.
Where you can use it
For the first time, Gemini Omni is generally available, backed by Google Cloud’s infrastructure and security, with Provisioned Throughput for workloads and an imperceptible SynthID watermark on outputs. Developers can prototype in Google AI Studio and build on the Gemini API, while businesses work through the Google Cloud console. Consumers get it through Flow, Google Vids and the Gemini app.
The Gemini app rollout starts with Google AI Ultra subscribers, and Google Ads gets the model for production. That distribution is the point: the same model a developer calls from the API also powers 15-second clips in Vids and scene direction in Flow. Google is not launching a standalone video tool; it is putting one video engine across its stack.
Our read
The 15-second number looks small, but it crosses a real threshold. Ten seconds is a clip; fifteen seconds is a scene, long enough for a setup, a turn and a payoff. That is why Google’s examples are all complete mini-stories, a try-on sequence and a pavilion-to-castle walk, rather than single shots.
The price cut does the same work from the other side: at about a dollar a clip, the cost of trying stops being a reason not to. When a tool is cheap enough to use casually, people use it casually, and casual use is where new formats are found.
There is a second signal here. Google is putting Omni 1.2 Flash everywhere at once, from the API to Vids to Ads, and leaning on Genie 4, its world model, as the showcase for what Omni’s consistency makes possible.
Video generation is no longer a feature Google demos; it is infrastructure Google ships, and the competition now has to match both the length and the price. That is a harder bar than a single demo, because it is a price, a length and a distribution network all at once.
Why It Matters
AI video has been stuck between two limits: clips too short to tell a story, and prices too high to experiment freely. Omni 1.2 Flash attacks both at once, and it does so as a generally available product with enterprise backing, not a preview.
For marketers, creators and developers, the practical effect is simple: a 15-second branded clip is now a routine generation, not a special project. Expect the volume of AI-made video ads to jump, and expect rivals to answer with longer clips and lower prices of their own.
Background
Gemini Omni is Google’s video generation and editing model, built on the Gemini family. Google has been expanding that family through 2026, including Gemini 4 Carbon for long-horizon engineering work and a top ranking for Gemini Carbon on an independent intelligence index. Omni 1.2 Flash is the first version of the video model to reach general availability.
Source: Google’s announcement
0 Comments