Google is making AI video generation more practical for developers, businesses, and content creators with the launch of Gemini Omni 1.2 Flash. Announced today, November 5, 2026, the latest version introduces native video generation of up to 15 seconds, improved consistency across frames, conversational editing, and a 30% price reduction compared with its predecessor.

The update is a meaningful step forward from Gemini Omni 1.1 Flash, which supported video generation of up to 10 seconds. With the new model, users can create longer clips in a single generation while maintaining better control over characters, products, environments, and camera movements.

Google is making Gemini Omni 1.2 Flash generally available through the Gemini API and Gemini for business, opening the door for production workloads rather than limiting the model to experimentation.

Gemini Omni 1.2 Flash makes video generation cheaper

One of the biggest changes is the reduction in video generation costs. Google has cut pricing by 30% compared with Gemini Omni 1.1 Flash, bringing the cost of generating 720p video down to $0.07 per second.

The updated workflow also gives creators more flexibility when producing videos. They can generate initial drafts at 360p or 720p, make adjustments through follow-up prompts, and upscale their preferred results to 1080p or 4K.

This approach can help reduce unnecessary regeneration costs, particularly when creating multiple versions of product videos, advertisements, and other commercial content.

Generate videos up to 15 seconds long

Gemini Omni 1.2 Flash increases the maximum native video generation duration from 10 seconds to 15 seconds. Users can also specify the length of a video in one-second increments, starting at three seconds.

That additional time gives creators more room to build complete scenes, product demonstrations, social media advertisements, and short explainers without stitching together several separately generated clips.

Google has also improved how the model follows visual references. Users can combine character images, product photographs, style references, motion clips of up to three seconds, and audio cues to guide video generation.

The model is designed to preserve important visual details throughout the clip, including faces, clothing, product packaging, typography, and brand colors. This could be particularly useful for businesses that need consistent product presentations across multiple camera angles.

Better consistency and conversational video editing

Visual consistency has been another challenge for AI video generators, especially when characters or objects move between scenes. Gemini Omni 1.2 Flash aims to address this with improved character, object, and environment consistency from the beginning to the end of a generated video.

Google says the model reduces unwanted visual changes and morphing during camera movements and transitions. It can also work with first and last frames to create smoother transitions between two reference points or extend existing footage.

Another important addition is multi-turn conversational editing. Instead of generating an entirely new video whenever something needs changing, users can refine the existing clip through follow-up instructions.

For example, creators can replace a product, change a background, adjust lighting or weather conditions, or modify the aspect ratio while retaining the original characters, performances, and camera framing.

This makes the editing process more flexible, particularly for marketing teams that need several variations of the same advertisement for different audiences and platforms.

More natural dialogue and cinematic camera movements

Google has also improved speech generation, lip-sync, and multi-speaker control. The model can produce more natural-sounding dialogue, clearer lip movements, and distinct voices for different characters.

These improvements are particularly relevant for animated scenes, explainers, and videos featuring multiple speakers. Google says the capabilities will be available across creative workflows involving Flow, Gemini for business, and Google Vids.

Camera control has received attention as well. Gemini Omni 1.2 Flash can follow more complex continuous movements, including sweeping first-person-view camera flights, dramatic zoom-outs, and orbital movements.

The goal is to reduce unwanted jump cuts and maintain a more consistent sense of scale throughout a scene, making generated footage look more cinematic.

Gemini Omni 1.2 Flash availability

Google is rolling out Gemini Omni 1.2 Flash across its developer and business platforms.

Developers can access the model through the Gemini API, prototype in Google AI Studio, and explore its documentation and pricing. Businesses can use it through Google Cloud and Gemini for business, with enterprise infrastructure and Provisioned Throughput available.

The model is also expanding into Google’s creative products:

• Google Vids: Users can generate 15-second videos at no additional cost with a Google account or Google Workspace subscription.
• Gemini app: 15-second video generation is rolling out to Google AI Ultra subscribers.
• Flow: Creators can direct scenes and make multi-turn edits.
• Google Ads: Advertisers can use the model to scale video campaign production.

Google says generated videos carry an imperceptible SynthID watermark. The company also acknowledges that development is ongoing in areas such as fast-moving sports scenes, rapid script changes, and complex physics.

With longer clips, improved reference following, conversational editing, and lower generation costs, Gemini Omni 1.2 Flash makes Google’s AI video technology more attractive for professional creative workflows. The combination of these improvements could be especially useful for businesses and creators producing video content at scale.