At a glance
Access: Google says: “Today, we’re rolling out Gemini Omni 1.2 Flash, and making it generally available across the Gemini API and Gemini for business.”
Pricing: Google says: “We have dropped the price 30% from Gemini Omni 1.1 making it available at $0.07 per second for 720p video.”
Official announcement: Gemini Omni 1.2 Flash
What shipped
Google released Gemini Omni 1.2 Flash on Nov 05 2026 and says it is generally available across the Gemini API and Gemini for business. The official post adds that, for the first time, Gemini Omni is generally available, backed by Google Cloud's enterprise-grade infrastructure and security.
The headline change is clip length. Videos can now be generated and edited natively up to 15 seconds, up from 10 seconds with Omni 1.1. Duration can be set in one-second increments from 3 to 15 seconds, which Google says leaves room for complete stories, product explainers or full social ad spots in a single generation.
Two operational details matter for production teams. Provisioned Throughput is available, and outputs carry an imperceptible SynthID watermark.
Key capabilities
- Multi-turn editing. You can refine the same clip turn by turn, such as swapping a product or background, changing the lighting or weather, or switching the aspect ratio, instead of starting over. Google says this works across full 15-second clips, so ad variants for different audiences and placements need no re-briefing or re-generating.
- Reference adherence. Generation can be guided by character, product or style images, short motion clips of up to three seconds, and audio cues. Google says faces, garments, packaging typography and brand colors stay consistent across camera angles.
- Consistency. Google describes stronger character, object and scene consistency from the first frame to the last, with less visual drift or morphing. It says this holds when interpolating between a first and last keyframe or extending a clip.
- Speech. Google lists more fluid, natural-sounding dialogue, clearer lip-sync and distinct multi-speaker conditioning.
- Camera control. Continuous moves such as FPV flythroughs, zoom-outs and orbits, with significantly fewer unwanted jump cuts.
- Cost path. Draft at 360p or 720p, fix a detail with a follow-up edit instead of re-generating the clip, then upscale the best results to 1080p or 4K.
Google also states what is still being worked on: fast sports action, rapid script changes and complex physics.
What it means for agent work
Most of these changes shape how an agent should plan and gate video jobs, not just how good a single clip looks.
Brief one clip, not a stitched set. With 15 seconds in one generation, a social ad spot or product explainer can be a single job. An agent can plan one brief per deliverable instead of joining 10-second segments.
Put the approval gate between draft and upscale. The 360p/720p draft tier and the 1080p/4K upscale step map onto a review queue. The agent drafts cheaply, a person approves, and only approved clips are upscaled. This is the same pattern covered in governed AI agents need approval queues and audit trails.
Make edits the default revision path. Because a follow-up edit replaces regeneration, an agent can keep each clip's edit history and produce variants from one approved base.
Route the known weak spots to human review. Sports action, rapid script changes and complex physics are the areas Google says are still improving. The consistency gains are Google's own descriptions, so test them on your own reference images before relying on them.
Keep provenance records. Because outputs carry a SynthID watermark, log the model and generation settings next to each approved asset.
Where Archon fits. Archon's Creative Studio is described on the site as "Cinematic video, brand visuals, and avatars at scale," and Archon is "model agnostic by design." A new video model is therefore a routing decision rather than a rebuild; see why enterprise AI should be model agnostic. To see how that works in practice, visit Archon Studio or Book a Consultation.
How to try it
Google says you can get started today as it rolls out across its products and developer platforms.
- Developers: prototype in Google AI Studio through the Gemini API, with interactive starter apps, developer documentation and pricing.
- Businesses: try it in the Google Cloud console and build enterprise solutions with the API.
- Google apps: direct scenes and multi-turn edits in Flow. Generate 15-second clips at no additional cost in Google Vids with a Google account or Google Workspace subscription. Create 15-second videos in the Gemini app, rolling out to Google AI Ultra subscribers. Scale video campaigns in Google Ads.
A sensible first run is a small batch of your own reference images at a fixed draft resolution. Check consistency there before committing to larger runs.