Google used the first week of November to push its generative video stack further into production workflows. On November 5, 2026, the company announced Gemini Omni 1.2 Flash on the Google blog, describing longer clips, stronger reference adherence, multi-turn conversational editing and a 30% price cut versus Omni 1.1 Flash. The model is generally available on the Gemini API and Google Cloud, the post says.
What is new in 1.2 Flash
According to Google’s announcement:
- Native video generation and editing up to 15 seconds, up from 10 seconds on Omni 1.1, with duration selectable in one-second steps from 3 to 15 seconds.
- Multi-turn conversational edits so creators can swap a product, background, lighting or aspect ratio without regenerating from scratch.
- Improved reference following for character, product and style images, plus short motion clips and audio cues.
- Pricing at $0.07 per second for 720p video — 30% lower than the prior Flash tier.
- Drafts at 360p or 720p, with upscaling of favored clips to 1080p or 4K as part of a cost-control workflow.
- Outputs carry an imperceptible SynthID watermark, and Provisioned Throughput is available for enterprise use on Google Cloud.
Why longer clips matter
Ten seconds is a bumper; fifteen seconds is a full social ad, a product spin or a short explainer beat. Google’s post emphasizes catalog try-ons, branded ads and continuous camera moves — FPV-style flythroughs, orbits and crash zoom-outs — with fewer unwanted jump cuts. For developers, the pitch is fewer full regenerations and more surgical follow-up prompts.
Where you can use it
Google says developers can prototype in Google AI Studio via the Gemini API; businesses can build in Google Cloud; consumers and Workspace users see 15-second generations rolling through Flow, Google Vids, the Gemini app (for AI Ultra subscribers, per the post) and Google Ads campaign tools. Availability still varies by product and subscription tier, so check the console or app notes before promising a client a feature that has not reached their account yet.
Limits Google itself flags
The blog notes ongoing work on fast sports action, rapid script changes and complex physics. That is a useful honesty check: Omni 1.2 Flash is positioned for scalable marketing and creative production, not as a finished replacement for every live-action shoot.
Practical takeaways for teams
- Budget with the per-second 720p rate and prefer edit-in-place over full rerolls.
- Keep brand reference packs (product shots, logos, character sheets) ready for consistency.
- Watermark awareness matters for disclosure policies even when the mark is imperceptible.
- Enterprise buyers should ask about Provisioned Throughput and data-handling terms before dumping customer footage into prompts.
Competitive context without hype
Google is not the only company selling short-form generative video, but Omni 1.2 Flash’s combination of length, price cut and edit-in-conversation is a clear product thesis: make iteration cheaper than regeneration. Teams evaluating tools should run the same brief through two vendors and score consistency, lip-sync and brand-color drift — the metrics Google itself emphasizes — instead of relying on launch-day demo reels alone.
Checklist before you put Omni on a client brief
Confirm your account has 1.2 Flash enabled, measure a 15-second branded test with your real product pack, and document how SynthID interacts with your disclosure policy. Only then update rate cards. Launch blogs are necessary reading; they are not a substitute for a pilot asset.
The bottom line
Google’s verified November 5 launch adds longer Omni clips, cheaper Flash pricing and conversational edits aimed at production scale. Read the company post for the source of record, then test on a single campaign asset before rewriting your whole video pipeline.
Image: Overhead desktop workspace (Unsplash) by Jeff Sheldon ugmonk via Wikimedia Commons, CC0 1.0
