Google has announced the general availability of Gemini Omni 1.2 Flash, the newest version of its multimodal video generation model. According to Google's official announcement, the update brings native 15-second video creation, multi-turn conversational editing, and improved reference adherence, while cutting prices by 30 percent. The model is rolling out today across the Gemini API, Google Cloud, Google Vids, and Gemini for business.
Anish Nangia, Product Manager at Google DeepMind, introduced the release in a blog post titled "Gemini Omni 1.2 Flash: Fast and cost-effective with longer clips," positioning it as a production-ready option for developers and businesses generating video content at scale. For more context on this story, see our ongoing AI news.
What's New in Gemini Omni 1.2 Flash
The headline capability is native support for 15-second video clips, a meaningful extension over previous versions of the Omni line. Google says the model also delivers stronger consistency across frames, which matters for professional projects where continuity between shots can make or break the final output.
The second major addition is multi-turn conversational editing. Instead of regenerating a clip from scratch when a detail is wrong, developers and creators can now refine videos through a sequence of conversational instructions. Google says this allows faster adjustments without starting over, which should reduce both iteration time and token costs for teams producing video in volume.
Google has also improved what it calls reference adherence — the model's ability to stick to the visual references and instructions a user provides. In practical terms, that means characters, styles, and scene elements guided by a reference image or description should stay more faithful across the generated clip.
Pricing Drops 30 Percent
Cost has been one of the main obstacles to production use of AI video generation, and Google is attacking it directly. The company reduced the price of Gemini Omni 1.2 Flash by 30 percent compared to the previous version, framing the model as a cost-effective option for high-volume video production.
The price cut arrives as competition in multimodal generation intensifies. OpenAI, Meta, and a wave of startups have all pushed into AI video over the past year, and developers increasingly choose models based on cost per clip as much as raw quality. Google's decision to pair longer clips with lower pricing suggests it wants Gemini Omni to be the default choice for commercial workloads rather than just experiments.
Where Developers Can Use It
Gemini Omni 1.2 Flash is generally available, which Google emphasizes means it is ready for production needs rather than experimental preview. Developers can access it through the Gemini API and Google Cloud, while business users can use it inside Google Vids and Gemini for business to produce video content for marketing, training, and communication workflows.
That distribution matters. Embedding the model directly into Workspace-adjacent tools like Google Vids gives Google a channel to enterprise customers that pure-play video generation startups have to fight for one contract at a time.
The Omni Line's Rapid Cadence
Gemini Omni 1.2 Flash arrives roughly ten weeks after Google introduced Gemini Omni 1.1 Flash in late August, and about five and a half months after the original Gemini Omni debuted in May. The steady cadence of roughly quarterly updates mirrors how quickly Google now iterates its frontier model families, with Flash variants positioned as the fast, affordable tier of each generation.
The release drew immediate attention from the developer community. The launch topped Hacker News on Friday morning, accumulating more than 170 points and over 130 comments within hours, with discussion focused on pricing, the 15-second limit, and how the model's editing workflow compares to alternatives.
What It Means for the AI Video Market
The update lands at a turbulent moment for AI companies. This week's stock market selloff, triggered in part by investor concerns about AI monetization, has put every major lab under pressure to show that frontier research translates into revenue. Cheaper, faster video generation is a direct answer to that question: lower prices expand the customer base, and production-ready status invites enterprise budgets that experiment-stage tools never see.
For developers, the practical takeaway is straightforward. If video generation workloads were previously constrained by clip length, price, or the pain of regenerating output after every change, Gemini Omni 1.2 Flash removes at least two of those three friction points today. Whether rivals respond with matching price cuts — as they have after previous Gemini Flash releases — will be one of the stories to watch in the coming weeks.
The consistency improvements deserve particular attention from teams that tried earlier video models and walked away. Google's announcement specifically targets "professional projects," acknowledging that frame-to-frame coherence — characters that look the same across shots, lighting that doesn't shift mid-scene — is what separates a demo from deliverable content. Combined with conversational editing, the workflow Google is describing looks less like a slot machine that regenerates clips and more like an iterative design tool.
Google says developers can start using the new capabilities immediately across the Gemini API, Google Vids, and Gemini for business.
---
Stay Ahead of AIGet the latest AI news, analysis, and breakthroughs — all in one place.
Read more AI news →
