Google drops Gemini Omni 1.2 Flash with 15-second video and a 30% price cut
Google just launched Gemini Omni 1.2 Flash, boosting native video generation to 15 seconds while slashing pricing by 30% to $0.07 per second. The update adds multi-turn conversational editing so you can tweak scenes without regenerating from scratch, alongside tighter reference consistency and realistic lip-sync. Google account users can now generate three 15-second clips every month in Google Vids for free.

Anyone who generates AI video knows the pain: you get a nearly perfect shot, but one prop in the corner is wrong, so you're forced to re-roll the entire prompt and pray the character's face doesn't melt.
Google just rolled out Gemini Omni 1.2 Flash across its Gemini API and cloud infrastructure to kill that exact loop. Native video length now reaches 15 seconds, pricing is down 30% to $0.07 per second for 720p footage, and you can finally edit clips conversationally without throwing out the whole generation.
Turn-by-turn tweaks without starting over
The most practical upgrade is the workflow itself. Instead of burning budget on high-res re-rolls, Google is encouraging teams to draft cheaply at 360p or 720p, fix what's broken in conversation, and upscale the winner to 1080p or 4K.
That conversational editing lets you swap backgrounds from an open pavilion to an oak forest or a stone castle while keeping character movements, camera tracks, and armor reflections intact. You can set durations from 3 to 15 seconds in one-second increments, giving social marketers enough runway for a full commercial spot without stitching clips together.
Visual consistency gets a major boost too. You can feed the model character photos, front-and-back clothing shots, and up to three seconds of motion references. It preserves garment drape, typography, and facial features across full 360-degree spins. Audio is sharper as well, with realistic multi-speaker routing and lip-sync across scene dialogue.
Google is rolling this out everywhere at once. Anyone with a personal Google account or Google Workspace subscription gets three free 15-second clips per month inside Google Vids at vids.new. Ultra subscribers get access in the Gemini app, while enterprise partners like Adobe, Figma Weave, and Monks are already wiring the API into their pipelines.
Google did note that fast athletic action and complex real-world physics are still works in progress.
Why it matters
We think Google is making the right call by optimizing for real studio budgets rather than just raw shock value.
Fifteen seconds is the sweet spot for modern digital ads and product explainers. Pushing native duration past 10 seconds means creators no longer have to splice multiple clips to build a basic social hook. Combining conversational turn-based edits with low-res drafting makes iterative client work actually feasible.
For the broader video AI market, this puts massive pricing pressure on standalone video startups. Delivering high-consistency video at $0.07 a second alongside enterprise-grade Cloud SLAs, SynthID watermarking, and free workspace tiers gives Google immense distribution leverage.
That said, the cracks still show whenever scenes demand rapid motion or intricate physics. For catalog spins, virtual try-ons, and polished corporate explainers, Omni 1.2 Flash is an easy recommendation. For cinematic action sequences, you'll still feel the guardrails.
Sources (4)
- blog.google(opens in a new tab)Originalblog.google
- blog.google(opens in a new tab)Reportblog.google
- x.com(opens in a new tab)Discussionx.com
- x.com(opens in a new tab)Discussionx.com