Google has launched Gemini Omni 1.2 Flash for general availability with the ability to generate video clips up to 15 seconds long at a 30% price reduction, supporting sequential conversational editing and upscaling to 4K resolution.
Gemini Omni 1.2 Flash rolled out with longer clips and 30% lower cost
Low-cost drafting, integrated conversational editing, and 4K resolution upscaling
Visual consistency with reference images and doubled developer usage limits
Deployment scope across Google platforms and early adoption by Adobe and Photoroom
Analytical reading of the strategic and economic dimensions of the event
Frequently asked questions
Gemini Omni 1.2 Flash rolled out with longer clips and 30% lower cost
Alphabet’s Google launched its updated AI video generation model, “Gemini Omni 1.2 Flash,” for general availability on Thursday, November 5, 2026. According to the official announcement from Google DeepMind, the new update enables the production of video clips up to 15 seconds long, up from the previous 10-second limit. It also allows users to edit and modify the video clip across multiple sequential conversational rounds while reducing financial costs by 30% compared to the previous version, “Gemini Omni 1.1.”
Google priced 720p video generation at $0.07 per second. According to the Google Cloud blog, the company is marketing this new model to serve categories of videos produced in large volumes by businesses and commercial enterprises, such as product showcase videos, virtual clothing try-on experiences, and digital advertisements.
Low-cost drafting, integrated conversational editing, and 4K resolution upscaling
Users can adjust the duration of the generated clip at any point between 3 seconds and 15 seconds in 1-second increments. Google states that this duration is fully sufficient for telling a complete story, providing a comprehensive product explanation, or building a complete social media advertisement in a single generation process. Tech publication The Decoder reported that the greatest financial savings stem from how users work with the model; they can create initial drafts in 360p or 720p resolution and fix any detail via a follow-up modification prompt rather than regenerating the entire clip from scratch, subsequently upscaling only the finalized, approved version to 1080p or 4K resolution, with other resolutions receiving comparable price cuts.
The conversational editing system operates efficiently across the full 15-second duration, allowing users to replace a specific product, change the background, adjust cinematic lighting, or swap the aspect ratio without starting over.
Visual consistency with reference images and doubled developer usage limits
Google emphasized that the updated model adheres with a higher degree of precision to attached reference images, preserving facial feature consistency, clothing details, and official brand colors across different camera angles. It also generates more natural spoken dialogue with clearer lip-syncing.
Developer release notes indicated that paid-tier rate limits have been doubled, and all API requests directed to the older 1.1 model will be automatically routed to the new 1.2 version starting December 3, 2026.
Deployment scope across Google platforms and early adoption by Adobe and Photoroom
Developers can now test the new model in Google AI Studio and build applications via the Gemini API, while enterprises can use it via Google Cloud. For general users, Google is adding the model to the Flow tool for directing scenes and executing multi-turn edits, to the Google Vids app where 15-second clips incur no extra cost, to the Gemini app for Google AI Ultra subscribers, and to Google Ads for visual advertising campaigns, complete with SynthID invisible watermarking on every output and enterprise-dedicated processing capacity available for purchase.
Regarding adopting companies, Adobe announced it will integrate the new update into its Firefly creative application. Matt Schوتين, senior director of products at Adobe, stated that it grants creators “longer outputs and stronger consistency with reference images.” Meanwhile, Photoroom CEO Matt Rouif noted that longer clips were “a key user request” and that general availability with service level guarantees means “we can rely on it in a real production environment for merchants and enterprise clients.”
Analytical reading of the strategic and economic dimensions of the event
The rollout of Gemini Omni 1.2 Flash for general availability reflects the focus of Google and Google DeepMind on transforming AI video generation technologies from dazzling visual demonstrations into highly cost-efficient commercial and enterprise production tools. Raising the maximum generated clip duration from 10 seconds to 15 seconds while cutting the price by 30% to $0.07 per second at 720p directly addresses the two most prominent obstacles faced by businesses and advertising agencies when using previous generations of generative video models.
The most significant practical value of this release lies in the interactive workflow based on multi-turn conversational editing. Designers no longer need to regenerate the entire video and waste computing budget when wanting to adjust lighting angles, swap scene backgrounds, or change aspect ratios. Instead, they can complete drafts at low resolution and upscale only the final version to 4K resolution while maintaining character feature consistency, brand colors, and lip-syncing. With doubled developer usage limits, integration across Google’s cloud and advertising ecosystem, and early adoption by companies like Adobe and Photoroom, Google is establishing a new competitive benchmark in the enterprise visual content production market.
Frequently asked questions
Question: What are the most prominent improvements in the Gemini Omni 1.2 Flash model compared to version 1.1?
Answer: The new model raises the maximum video duration from 10 seconds to 15 seconds, supports conversational editing across multiple rounds, and reduces costs by 30%, bringing the price per second at 720p down to $0.07.
Question: How does the model help businesses and designers lower video production costs?
Answer: It enables the creation of fast drafts at 360p or 720p and conversational modification of details like lighting, backgrounds, and aspect ratios without regenerating from scratch, upscaling only the final version to 1080p or 4K resolution.
Question: When will requests for the older 1.1 version be automatically transferred to version 1.2?
Answer: Developer notes stated that automatic routing from version 1.1 to version 1.2 will begin on December 3, 2026, alongside doubled usage limits for paid tiers.
Question: What are the key platforms and companies that have adopted the Gemini Omni 1.2 Flash model?
Answer: It is available via Google AI Studio, Google Cloud, Flow, Google Vids, Google Ads, and the Gemini app for Ultra subscribers. Adobe and Photoroom have also announced integrating it into their services.