Gemini Omni 1.2 Flash: Google's Video Model Gets Longer Clips and a 30% Price Cut
Gemini Omni 1.2 Flash makes 15-second AI videos, edits clips over several turns and costs $0.07 a second at 720p. Here's what Gemini Omni means for creators.

Google released Gemini Omni 1.2 Flash on 5 November 2026, an updated version of its video generation and editing model. In a blog post by Anish Nangia, a product manager at Google DeepMind, the company said the new model makes native video clips of up to 15 seconds (up from 10 seconds in Omni 1.1), lets you edit the same clip over several conversational turns, and costs 30% less than the previous version, at $0.07 per second for 720p video. It is generally available on the Gemini API and Google Cloud, and it is also turning up in Flow, Google Vids, Google Ads and, for Google AI Ultra subscribers, the Gemini app.
This article covers what Google announced, why the longer clip length and lower price matter more than they first appear, the details on pricing and where you can use it, and what Gemini Omni means for creators, small businesses and developers, including what a clip actually costs in naira.
What Happened
Gemini Omni is Google's family of models that generate and edit video, with audio, from text, images and other references. Version 1.2 Flash is the newest release, and Google is pitching it as a model for "producing videos at scale" rather than one-off showpieces.
According to Google's announcement, the headline changes are:
Longer clips. Native generation and editing up to 15 seconds, set in one-second steps anywhere from 3 to 15 seconds. Omni 1.1 topped out at 10 seconds.
Multi-turn editing. You can keep refining the same clip in conversation (swap a product, change the background, adjust lighting or weather, switch the aspect ratio) without regenerating it from scratch. Google says this works across the full 15 seconds.
Better reference adherence. The model can follow character, product or style images, short motion clips of up to 3 seconds, and audio cues. Google says faces, garments, packaging text and brand colours stay more consistent across camera angles.
More natural speech. Clearer lip-sync, more natural dialogue, and the ability to assign different voices to specific characters in a scene.
A 30% price cut. $0.07 per second for 720p video through the API.
Google also listed better camera control (orbits, zoom-outs, first-person flythroughs) with fewer unwanted jump cuts, and better consistency from the first frame to the last, with less of the "morphing" that makes AI video look off.
Why It Matters
Five extra seconds sounds minor. It isn't. Most short-form video on Instagram, TikTok and YouTube Shorts is built from shots that run a few seconds each, and a full ad or product teaser often runs 15 to 30 seconds. At a 10-second cap, you had to stitch two or three AI clips together for almost anything useful, and the joins were where characters changed faces and products changed colour. A 15-second ceiling means a complete short ad, a product spin or a single scene can come out as one clip, with one consistent look.
Multi-turn editing matters for a similar reason. With most video models, a small mistake (the wrong shirt colour, a sign with garbled text, a background you don't like) meant starting again and paying again, with no guarantee the next attempt keeps what you liked about the first. Being able to say "keep everything, change the background to a Lagos street at dusk" is closer to how people actually edit. If it works as described, it cuts both cost and frustration.
The price cut is the third piece. Video generation has been one of the most expensive things you can do with AI, and price per second has kept many small creators and businesses on free tiers with heavy limits. A 30% drop doesn't make it cheap, but it moves short clips into the range where testing several versions of an ad is realistic for a small team.
There's also a competitive angle. OpenAI used to have Sora as its consumer video app, and this week it shipped animated GIFs in ChatGPT instead (we cover that in ChatGPT Images 2.6 brings GIFs and faster image generation). Google is pushing in the other direction, with full video, sound and editing built into tools people already use for work.
The Details
Pricing
Google's post gives one price: $0.07 per second of 720p video, which it says is 30% lower than Omni 1.1. Working backwards, that implies Omni 1.1 cost about $0.10 per second at the same resolution. Google did not list separate prices for 360p drafts or for upscaling to 1080p or 4K in the announcement, so check the Gemini API pricing page before you budget for those.
Here is what the 720p rate works out to for common clip lengths:
Clip length (720p) | Cost in dollars | Rough cost in naira |
|---|---|---|
5 seconds | $0.35 | about ₦470 to ₦490 |
8 seconds | $0.56 | about ₦755 to ₦785 |
10 seconds | $0.70 | about ₦945 to ₦980 |
15 seconds | $1.05 | about ₦1,420 to ₦1,470 |
Naira figures use roughly ₦1,350 to ₦1,400 to the dollar (an estimate; check the current rate). The official NFEM rate was about ₦1,348 and the parallel market about ₦1,390 to ₦1,400 on 6 November 2026. Your card or payment provider may add fees on top.
Remember that you pay for every generation and, most likely, every edit pass, not just the clip you keep. If it takes four attempts to get a 15-second ad right, that's about $4.20, or roughly ₦5,700 to ₦5,900.
The suggested workflow
Google recommends a specific way of working to keep costs down: draft at 360p or 720p, fix details with follow-up edits rather than full regenerations, and only upscale the final version to 1080p or 4K. This is sensible advice for any AI video tool. Low-resolution drafts are where you check the idea, the timing and the composition. High resolution is for the version you'll actually post.
Where you can use it
According to Google, Gemini Omni 1.2 Flash is available in these places:
Gemini API: generally available, with prototyping in Google AI Studio (model id
gemini-omni-1.2-flash).Google Cloud: generally available through Vertex AI Studio and the enterprise API, with Provisioned Throughput for businesses that need guaranteed capacity.
Flow: Google's filmmaking tool, for directing scenes and making multi-turn edits.
Google Vids: 15-second clips at no additional cost with a Google account or Google Workspace subscription.
Gemini app: 15-second videos rolling out to Google AI Ultra subscribers.
Google Ads: for scaling video campaigns through Asset Studio.
The Google Vids line is the one most readers should notice. If you already use Google Workspace, or a personal Google account with access to Vids, you can try the model's 15-second clips without paying per second.
Safety and watermarking
Every output carries an invisible SynthID watermark, Google's system for marking AI-generated media so it can be detected later. That won't stop misuse on its own, but it means Google's tools can identify a clip as AI-made even after it's been shared. If you want to know how to spot AI media yourself, the habits you'd use to tell if an image is AI generated apply to video too.
What Google says it still can't do well
Google was unusually direct about weak spots. It said it is still working to improve fast sports action, rapid script changes and complex physics. In practice, that means a footballer dribbling past three defenders, a scene where dialogue changes quickly, or anything involving liquids, collisions or objects falling naturally may still look wrong. Plan around those limits rather than discovering them after you've spent money.
It's also worth noting that every example in Google's post was a Google-made demo. The company said businesses and startups are already using the model but didn't name any. Independent testing will tell us how the consistency and editing claims hold up.
What It Means for You
If you're a creator
Gemini Omni 1.2 Flash is most useful for B-roll, product shots, intros, transitions and short scenes that would be expensive or impossible to film yourself. It's less useful as a replacement for your face, your voice and your opinion, which is still what makes people follow an account. There's a reason why AI videos fail to go viral: technically impressive clips still flop without a hook, a point of view and visuals that feel your own.
If you run a faceless channel, the longer clips and character consistency help directly, because keeping the same character across a whole video has been the hardest part. Our faceless videos with AI guide shows how to build that kind of channel without it looking like every other AI account.
Whatever you generate, you still need to cut it together, add captions and fit it to the platform. A content editor such as Nexframe Studio can handle that last stretch, and once your clip is ready, our workflow for Instagram captions with AI helps you write the post that goes with it in your own voice.
If you run a small business
The reference features are aimed squarely at you. Google says the model keeps garments, packaging text and brand colours consistent from different angles, which is what you need for a product spin, a "virtual try-on" style clip, or a short ad for a new arrival. A 15-second clip is long enough for a full product teaser.
Start small. Pick one product, write a clear brief (what it is, who it's for, the mood, the setting, the aspect ratio), draft at low resolution, edit in a couple of turns, then upscale. Check every frame for errors in the product itself: wrong logo, extra buttons, misspelled text. Customers notice, and a fake-looking ad can do more harm than no ad.
Also be honest. If a clip shows a "customer" who doesn't exist wearing your product, don't present it as a real testimonial.
If you're a developer
General availability on the Gemini API and Vertex AI means you can build Gemini Omni into products, not just experiment. The multi-turn editing is the interesting part for app builders: you can let users refine a clip step by step rather than rerolling. Budget carefully, though. At $0.07 per second, a feature that generates a 15-second clip for every user action can get expensive fast, so add limits, caching of drafts and clear user feedback before an expensive upscale.
If you're in Nigeria
The most practical route for most Nigerian creators and businesses is Google Vids, because Google says 15-second clips there come at no additional cost with a Google account or Workspace subscription. That avoids dollar billing entirely, which matters when card limits and exchange rates make every API charge a bit unpredictable.
If you do pay through the API, use the table above as a guide. A 15-second 720p clip costs about ₦1,420 to ₦1,470 at today's rough rates. A small brand making 30 finished clips a month, with an average of three generations per final clip, would spend about $94.50, or roughly ₦128,000 to ₦132,000 (an estimate; check the current rate and any card fees). That's real money, so the low-resolution draft habit isn't optional.
Speech quality for Nigerian accents, Pidgin and local languages isn't something Google addressed. Test a short clip before committing to a campaign that depends on dialogue.
What to Watch Next
Independent tests: Whether character and product consistency actually hold across a full 15 seconds, and whether multi-turn edits really keep the rest of the clip unchanged.
Gemini app rollout: Google says 15-second videos are rolling out to Google AI Ultra subscribers first. Watch for whether lower tiers get access, and with what limits.
Upscale and draft pricing: The announcement only gives the 720p price. The cost of 1080p and 4K upscales will decide how affordable final outputs really are.
Competitor responses: OpenAI has moved into GIFs in ChatGPT, and other video model makers will likely answer on clip length and price.
Ad policies: As AI video spreads through Google Ads, expect more rules on labelling and on what counts as misleading content.
Final Thoughts
Gemini Omni 1.2 Flash is an incremental release with practical consequences. Fifteen-second clips cover a full short ad in one generation, multi-turn editing cuts down on expensive rerolls, and the 30% price drop makes testing several versions realistic for smaller teams. Google's own list of weak spots (fast action, rapid dialogue changes, complex physics) is a useful reminder that this is still a tool you direct, not one you trust blindly. Try it in Google Vids first, draft at low resolution, and spend your money on the clips that already work.
Frequently Asked Questions
What is Gemini Omni 1.2 Flash?
It's Google's latest video generation and editing model, released on 5 November 2026. It creates video with audio from text, images, short motion clips and audio cues, generates clips of 3 to 15 seconds, and lets you edit the same clip over several conversational turns. Google says it is generally available on the Gemini API and Google Cloud.
How much does Gemini Omni 1.2 Flash cost?
Google says it costs $0.07 per second for 720p video through the API, 30% less than Omni 1.1. A 15-second 720p clip costs $1.05, or about ₦1,420 to ₦1,470 at roughly ₦1,350 to ₦1,400 to the dollar (an estimate; check the current rate). Google says 15-second clips in Google Vids come at no additional cost with a Google account or Workspace subscription.
How long can Gemini Omni videos be?
Up to 15 seconds per clip, set in one-second steps from 3 to 15 seconds. The previous version, Omni 1.1, was limited to 10 seconds. For longer videos, you still need to combine several clips in an editor.
Can I use Gemini Omni in the Gemini app?
Google says 15-second videos are rolling out in the Gemini app to Google AI Ultra subscribers. If you're not on Ultra, Google Vids and Flow are the easier consumer routes, and developers can use Google AI Studio.
Are Gemini Omni videos watermarked?
Yes. Google says every output carries an imperceptible SynthID watermark that identifies it as AI-generated. It isn't visible to viewers, but Google's detection tools can read it.
Can I use Gemini Omni from Nigeria?
Google AI Studio and Google Cloud are generally usable from Nigeria, but billing in dollars can be awkward with local cards. Google Vids, which Google says includes 15-second clips at no additional cost with a Google account or Workspace subscription, is the simplest place to start. Check availability in your own account, since rollouts can vary by region.
Comments (0)
Sign in to post a comment.
- Be the first to comment.


