Google unveils Gemini Omni 1.1 Flash, an updated multimodal AI model aimed at improving how creators make and refine AI-generated videos. The model focuses on adding more controllable tools during generation, including the ability to work with longer video context to extend scenes rather than starting from scratch.
Across reports, the new capabilities are presented as practical workflow enhancements. Sources say Gemini Omni 1.1 Flash can connect two user-specified frames and can use short reference video clips while generating new footage. Google also includes options to preview video outputs at lower resolution for faster iteration before producing higher-quality results, including 1080p and 4K.
NDTV emphasizes additional creator-focused controls and notes that the model supports a 40-second extension workflow, along with lower-cost, speed-focused previews that can precede final rendering. TechGenyz highlights scene extension as a key feature and cites Google’s description that the model can analyze preceding footage (with specifics mentioned in its report). Both sources agree that access is provided via Google’s developer platforms, with subscriber availability through Google Flow and related Gemini features.
