Google releases Gemini Omni 1.2 Flash: 15-second video, 30% lower price, free clips in Google Vids
On November 5, 2026 Google made its Gemini Omni 1.2 Flash video model generally available. Clips now run up to 15 seconds instead of the 10 allowed by Omni 1.1, and pricing is 30% below Omni 1.1. The model is live in the Gemini API, Google Cloud, Flow and Google Vids, where anyone with a Google account or Workspace subscription can make up to three 15-second clips a month at no cost.
Omni 1.2 Flash arrives with longer clips and a lower price
Google introduced Gemini Omni 1.2 Flash in an announcement dated November 5, 2026, written by Anish Nangia, who is presented as a product manager at Google DeepMind. Google frames the model as a fast, cost-effective option for generating and editing video. The headline change is length: the model can generate and edit video natively up to 15 seconds, where Omni 1.1 stopped at 10. Duration can be set in one-second steps anywhere from 3 to 15 seconds.
Google says it cut pricing 30% against Omni 1.1 and publishes a per-second rate for 720p video; we leave the figure out here, and the developer pricing page is the place to check current rates. The company also says Gemini Omni is generally available for the first time, backed by Google Cloud's enterprise infrastructure and security. A second announcement covers Google Vids, which now runs on the same model.
Reference fidelity, conversational editing and more natural speech lead the list
The model accepts character, product or style images, motion clips of up to three seconds, and audio cues as joint guidance. Google says faces, garments, packaging type and the brand palette hold steady as the camera angle changes, and that existing product photography can be turned into clothing try-on clips, catalog rotations and branded video ads. Visual drift and morphing are reduced, which also helps when bridging a first and last frame or extending a clip. On speech, Google lists more fluid dialogue, clearer lip-sync, and conditioning that routes each voice to the right character in multi-speaker scenes. For camera work it points to continuous moves such as FPV flythroughs, zoom-outs and orbits with fewer unwanted cuts.
Editing happens turn by turn: swapping a product or background, changing lighting or weather, or switching aspect ratio no longer means regenerating the clip. Google says this works on full 15-second clips and lets teams turn out ad versions aimed at different audiences and placements. It also suggests a cost-conscious workflow: draft at 360p or 720p, fix details with a follow-up edit, then upscale the best result to 1080p or 4K. Outputs carry an imperceptible SynthID watermark. Google acknowledges that it is still working on rapid sports movement, quick script swaps and intricate physics.
One model now shows up everywhere from developer tools to Vids
Google lists where the model can be used. Developers can prototype in the Gemini API through Google AI Studio; businesses can use the Google Cloud console, and Provisioned Throughput is offered. Flow supports scene direction and multi-turn edits. In the Gemini app, 15-second videos are rolling out to Google AI Ultra subscribers. The post also names Google Ads for scaling video campaigns, without saying which Ads surface it means. We covered Omni's arrival in Google Ads on August 26.
The second post reaches the widest audience. Google Vids has moved to this model, and a Google account or a Workspace subscription is all it takes to produce as many as three 15-second videos each month, free of charge. The uses Google names for Vids include single-take product promos without a cut, videos built from uploaded reference images with consistent characters and scenes, and more natural narration. The practical upshot: trying a short video no longer requires a camera crew or a studio setup, only an account, within the three-clips-a-month Vids limit.
The posts say nothing about Türkiye, so this section is commentary
Neither Google post mentions Türkiye, and neither draws a country, region or language line around access. This section is therefore not a fact from the sources but plainly UNALSOFT commentary.
The unknowns, one by one. The posts do not say when or to which accounts the Google AI Ultra rollout in the Gemini app will reach Türkiye. They do not say whether the three free monthly clips in Vids apply to accounts in Türkiye. There is no statement on Turkish speech or lip-sync support, and nothing on whether businesses in Türkiye will get the model in Google Ads. Before planning around it, test it in your own account.
The UNALSOFT view
We read this as a tool update that shortens the time and cost of testing a short ad. Fifteen seconds is enough for a product promo or a social placement in one piece, and reference images plus conversational editing let one product spawn several variants without rewriting the brief. Still, the tool is not the work: brand consistency, the message and the call on which clip actually runs belong to a person inside a production line. On the AI ad films side we use models like this as one part of that line. Since the posts say nothing about Türkiye, this piece does not suggest any local access or legal conclusion.
Want to test short video ads faster?
A short conversation is enough to work out which video workflow fits your brand.