Google ships Gemini Omni 1.2 Flash with 15-second clips, multi-turn editing and 30% price cut

What Google released

Google has released Gemini Omni 1.2 Flash, a new version of its video generation model, and made it generally available through the Gemini API and Gemini for business. According to the Google AI blog, the update introduces native 15-second video creation, multi-turn conversational editing and improved reference following, while cutting the price by 30% versus Gemini Omni 1.1 to $0.07 per second for 720p video.

The pricing story is not just the headline rate. Google describes a workflow designed to keep costs down at scale: draft quickly at 360p or 720p, fix individual details with a follow-up edit rather than regenerating the whole clip, then upscale only the best results to 1080p or 4K.

Longer clips with one-second granularity

Clip length grows from 10 seconds in the previous version to a maximum of 15 seconds, and creators can now set duration in one-second increments between 3 and 15 seconds. Google positions this as enough room to tell complete stories, build product explainers or produce full social ad spots in a single generation.

Stronger consistency and reference handling

The model can be guided with a mix of character, product or style images, motion clips of up to three seconds and audio cues — and Google says it holds those details across the whole output. Faces, garments, packaging typography and brand colors are meant to stay consistent across camera angles, which Google suggests makes it easier to turn existing product photography or character sheets into virtual try-on clips, catalog spins and branded video ads.

Google also says drift is reduced: characters, objects and scenes hold together from the first frame to the last, including when interpolating between keyframes or extending a clip into a longer sequence. The company's own comparison shows Omni 1.1 Flash matching a reference image less closely than 1.2 on a layered animation prompt.

Editing in conversation

Rather than starting over when something needs to change, users can refine the same clip turn by turn — swapping a product or background, adjusting lighting or weather, or switching the aspect ratio. Multi-turn editing works across full 15-second clips, which Google frames as a way to produce ad variants for different audiences and placements without re-briefing or regenerating.

Better speech and camera work

Dialogue is described as more fluid and natural, with clearer lip-sync and multi-speaker conditioning that routes each sampled voice to the right character. These improvements surface in Flow for directing multi-character scenes, in Gemini for business video explainers, and in Google Vids for polished videos with avatars, music and narration.

Camera direction also gets more control. Google says the model handles continuous moves such as FPV flythroughs, dramatic zoom-outs and orbits with fewer unwanted jump cuts and steadier scale; one example describes a continuous drone-style flight through the Pantheon with no cuts.

Where to get it

This is the first time Gemini Omni is generally available, backed by Google Cloud's enterprise-grade infrastructure and security, with Provisioned Throughput offered and an imperceptible SynthID watermark applied to outputs.

Availability is broad: developers can prototype in Google AI Studio and build with the Gemini API; businesses can work through the Google Cloud console; Google Vids offers 15-second clips at no additional cost with a Google account or Workspace subscription; the Gemini app gets 15-second video rolling out to Google AI Ultra subscribers; and Google Ads customers can use it to scale video campaigns.

Google is upfront about remaining gaps, noting that development continues on fast sports action, rapid script changes and complex physics.

Why it matters

Fifteen seconds is the length of a standard social ad, and per-second pricing at $0.07 makes high-volume output cheaper just as brands begin testing generative video in real campaigns. The bigger shift may be multi-turn editing: iterating on a single clip instead of re-rolling it changes both the cost model and the creative workflow, bringing video generation closer to how designers already work. Combined with enterprise availability, Provisioned Throughput and SynthID watermarking, Google is clearly aiming this release at commercial buyers rather than hobbyists — and applying pressure on rivals in a fast-moving text-to-video market.