Google Releases Gemini Omni 1.1 Flash AI Model for Video Generation
Google announced on August 27th the release of its Gemini Omni 1.1 Flash AI model, capable of generating videos with up to 4K resolution. The model introduces new features for extending and editing video content.
Google unveiled its Gemini Omni 1.1 Flash AI model on August 27th, a tool designed for video generation that can produce content up to 4K resolution. This release expands the company's artificial intelligence capabilities in media creation.
The Gemini Omni 1.1 Flash model includes several key enhancements. Users can now extend existing videos by seamlessly adding subsequent scenes, allowing for a cumulative video length of up to 40 seconds. The model also permits the specification of start and end frames, facilitating smooth transitions and controlled camera movements.
Additionally, the model offers a faster and more cost-effective option for generating previews. Creating 360p previews can be up to 60% faster and cost one-third of the standard rate, suitable for rapid prototyping and script iteration. It also supports the use of reference video clips, up to three seconds long, to maintain consistency in context and character.
Pricing for the service varies by resolution: 360p video generation costs $0.03 per second, 720p at $0.1 per second, 1080p at $0.15 per second, and 4K at $0.3 per second. This advancement positions Google to compete further in the AI-driven video production market.