Google Gemini Omni 1.2 Flash has been released, with support for video clips up to 15 seconds long.

Google DeepMind has introduced Gemini Omni 1.2 Flash. This latest video generation model supports the native creation of 15-second videos and features new editing capabilities. According to Google, the new model is 30% more affordable than its predecessor, Gemini Omni 1.1. Users can generate 360p or 720p video content at a cost of $0.07 per second. The model supports the generation and editing of continuous video clips up to 15 seconds in length, as well as conversational editing.

Google Launches Gemini Omni 1.2 Flash
On Thursday, Google announced the release of Gemini Omni 1.2 Flash via a blog post. Developers can access the model through the Gemini API and begin using it via Google AI Studio. The company is also making the model available to enterprise customers through Google Cloud.

Generating 720p video with Gemini Omni 1.2 Flash costs $0.07 per second—a 30% reduction compared to Gemini Omni 1.1. Google states that users can generate videos in 360p or 720p and perform subsequent edits—rather than regenerating the entire clip—before upscaling the final output to 1080p or 4K.

Gemini Omni 1.2 Flash allows users to natively generate and edit videos up to 15 seconds long—an increase from the 10-second limit supported by Gemini Omni 1.1. Users can specify the video duration in 1-second increments, ranging from 3 to 15 seconds. The new version also enables creators to guide video generation by combining images (such as people, products, or specific styles), short video clips (up to 3 seconds), and audio prompts. It is designed to maintain consistency across different camera angles—ensuring that facial features, clothing, text on packaging, and brand colors remain uniform.

Gemini Omni 1.2 Flash reportedly offers improved frame-to-frame consistency. Users can swap out products or backgrounds, alter lighting or weather conditions, or adjust the aspect ratio without needing to regenerate the entire video. Multi-turn conversational editing is also supported for videos up to 15 seconds long, allowing creators to produce multiple versions of an advertisement without having to start from scratch.

Google has also enhanced the model’s audio generation capabilities. Gemini Omni 1.2 Flash is said to produce more natural, fluid dialogue, achieve more precise lip-syncing, and deliver significantly better multi-channel audio. These improvements are set to benefit users creating videos via Flow, Gemini for Business, and Google Vids. Furthermore, the model supports complex camera movements, including first-person view (FPV) style fly-throughs, zooms, and orbiting shots.

Google states that Gemini Omni 1.2 Flash is designed for developers, enterprises, and content creators who need to produce video at scale. Developers can begin using the model via Google AI Studio and the Gemini API documentation, while enterprise customers can access it through the Google Cloud console. Google has confirmed that as the rollout expands, users will be able to explore its video generation capabilities through Google applications.

Leave a Comment