Gemini Omni 1.2 Flash: 15-Second Video at $0.07/sec, 30% Cheaper
By Ali Al-Montaser — Senior AI Technical Editor
Google DeepMind launches Gemini Omni 1.2 Flash, enabling native 15-second video generation with multi-turn conversational editing. Priced at $0.07 per second for 720p—a 30% reduction from version 1.1—it is immediately available via Gemini API, Google Cloud, Google Vids, and Gemini for business. The model enhances character and product consistency while reducing overall costs through a tiered workflow from 360p drafts to 4K upgrades.
Executive Overview
Google DeepMind has announced the launch of Gemini Omni 1.2 Flash, a native video generation model that supports clips up to 15 seconds and multi-turn conversational editing. Priced at $0.07 per second for 720p resolution—a 30% reduction from version 1.1—the model is immediately available for general use via Gemini API, Google Cloud, Google Vids, and Gemini for business. This release marks a significant step in making high-quality AI video generation more accessible and cost-effective for enterprises and creators.
📊 Official Technical Specifications & Data Sheet
| Technical Aspect | Confirmed Official Data |
|---|---|
| 💰 Pricing & Usage Cost | $0.07 per second of 720p video — 30% lower than Gemini Omni 1.1. Cost for a 15-second clip ≈ $1.05. Tiered workflow: 360p/720p draft then upgrade to 1080p or 4K. |
| 🌐 Platforms & Immediate Availability | Generally available (GA) via Gemini API, Google Cloud, Google Vids, Gemini for business, Google AI Studio, and Flow. Supports Provisioned Throughput. |
| ⚡ Performance & Speed Metrics | Native generation up to 15 seconds (50% increase from 10 seconds in 1.1). Duration adjustable in one-second increments from 3 to 15. Improved reference adherence and character consistency across shots. Reduced unwanted camera motion jumps. |
| 🛡️ Security & Tamper Resistance | Google Cloud enterprise-grade security infrastructure, invisible SynthID watermark on all outputs, and Provisioned Throughput support for production. |
| 🧠 Context Window | Token context window size not disclosed in official statement; supports multi-turn context for conversational editing on full 15-second clips, with visual references (character, product, and style images), motion clips up to 3 seconds, and audio cues. |
| 🌍 Arabic Language & Regional Support | No explicit Arabic support mentioned in the statement; model supports multi-speaker dialogue and voice translation, enabling Arabic content production via Google Vids and Gemini for business. Google Cloud availability includes Middle East regions. |
Deep-Dive Features & Architecture
Gemini Omni 1.2 Flash introduces substantial upgrades to its functional architecture: native generation extended from 10 to 15 seconds, and duration adjustable in one-second increments between 3 and 15 seconds, allowing for full social ads or product explainers in a single generation. Most importantly, the model offers multi-turn conversational editing that lets users modify the same clip through successive prompts—switching background, product, lighting, or aspect ratio—without full regeneration, reducing overall cost and accelerating production.
For reference adherence, the model supports integration of character, product, and style images with short motion clips up to 3 seconds and audio cues, maintaining consistency of faces, clothing, packaging typography, and brand colors across camera angles. In an official test, the model generated a 15-second "virtual try-on" sequence from a single model image and front/back images of three outfits, preserving the model's identity and details of cut and folds. The model also improved multi-speaker dialogue, voice translation, and comedic timing in 15-second scenes.
On cinematic control, the model supports complex continuous camera movements (FPV, zoom out, orbits) with fewer jumps and higher scale accuracy. In a demo, it produced a continuous FPV drone flight through the Pantheon courtyard to the oculus without any cuts. The workflow includes cost reduction via fast 360p or 720p drafts followed by upgrades to 1080p or 4K, with detail fixes through subsequent editing rather than regeneration.
Benchmark & Competitive Performance
Compared to Gemini Omni 1.1 Flash, version 1.2 offers a 50% increase in maximum clip length (from 10 to 15 seconds), a 30% price reduction to $0.07 per second, and tangible improvements in visual reference adherence and character consistency across shots. In a comparative test of a five-layer studio exhibition poster, version 1.1 failed to accurately match the reference image, while 1.2 built the poster layer by layer, preserving materials, shadows, and typographic details through the final frame. The model also demonstrated improved...
Industry Impact & Enterprise Adoption
The launch of Gemini Omni 1.2 Flash is poised to disrupt the AI video generation market by significantly lowering the cost barrier for high-quality, longer-form video content. Enterprises can now produce professional-grade 15-second clips for social media, product demos, and virtual try-ons at a fraction of previous costs. The availability across Google Cloud and Gemini for business, combined with Provisioned Throughput and SynthID watermarking, ensures that businesses can scale production securely and reliably. The tiered workflow from low-resolution drafts to 4K upgrades further optimizes budgets, making advanced video generation accessible to a broader range of organizations.
Conclusion
Gemini Omni 1.2 Flash represents a major leap in AI video generation, combining longer duration, conversational editing, and a 30% price cut to $0.07 per second. With immediate availability on major Google platforms and enterprise-grade security, it sets a new standard for cost-effective, high-quality video production. As businesses increasingly adopt AI-driven content creation, this model is well-positioned to become a cornerstone of enterprise video workflows.
Media Source: Google AI Blog | البيان الرسمي للشركة: المصدر الأصلي | Fact Verification & Analysis: AI Tools Oasis
Frequently Asked Questions
The official price is $0.07 per second of 720p video, a 30% decrease from Gemini Omni 1.1. Generating a 15-second video costs approximately $1.05.
The model supports native generation and editing of videos up to 15 seconds, up from 10 seconds in version 1.1, with duration adjustable in one-second increments from 3 to 15 seconds.
It is generally available immediately via Gemini API, Google Cloud, Google Vids, and Gemini for business, with Provisioned Throughput support and invisible SynthID watermarking on outputs.
The official announcement does not explicitly mention Arabic language support for the model, but it supports multi-speaker dialogue and voice translation, enabling Arabic content production via Google Vids and Gemini for business.
1.2 Flash offers 15-second videos instead of 10, multi-turn conversational editing, better reference adherence, a 30% lower price at $0.07/sec, and higher consistency of characters and products across shots.
AI Tools Oasis Team
Bringing you the latest news and analysis in the world of Artificial Intelligence with accuracy and credibility. Follow us for all updates.
