The launch of Gemini Omni 1.2 Flash by Google DeepMind introduces 15-second video production, improved consistency, and faster generation at a reduced cost. Available through the Gemini API, Google Vids, and Gemini for business, this update offers enhanced video creation capabilities for professional use.
The Omni 1.2 Flash version, now fully accessible across the Gemini API and business platforms, introduces features like native 15-second video creation and multi-turn conversational editing. The update improves reference adherence and reduces costs by 30%, with a new price of $0.07 per second for 720p video, a reduction from the previous version.
Key features of Gemini Omni 1.2 Flash include:
- Flexible Video Lengths: Videos can now be generated and edited up to 15 seconds, an increase from 10 seconds in the prior version. Users can specify durations from 3 to 15 seconds, providing versatility for storytelling and advertising.
- Enhanced Reference Adherence: Users can create videos using character, product, or style images, along with short motion clips and audio cues. This ensures consistency in details like faces, garments, and brand colors throughout the video.
- Consistent Visuals: The update improves consistency from the first to the last frame, minimizing visual drift during transitions and maintaining details across shots.
- Efficient Editing: Multi-turn conversational editing allows changes in products, backgrounds, lighting, or aspect ratios without starting over, ideal for creating diverse ad variants.
- Improved Dialogue and Motion: The update supports natural-sounding dialogue, clearer lip-sync, and precise multi-speaker coordination, enhancing the quality of multi-character videos.
- Advanced Camera Movements: Users can apply complex, fluid camera motions with fewer unwanted cuts, achieving a cinematic look in videos.
Businesses and creatives are already integrating Gemini Omni 1.2 Flash into their workflows. With enterprise-grade infrastructure and security from Google Cloud, the platform supports creating videos at scale. It also features a nearly invisible SynthID watermark for outputs. Development continues to enhance features like fast-paced sports action and complex physics simulations.
Users can begin utilizing Gemini Omni 1.2 Flash today across various Google products and platforms.