Google has introduced Gemini Omni Flash, the first model in its new Omni family, designed for creating and editing high-quality videos from text, image, video, or audio references. The company has opened access to the model for developers.
Key capabilities highlighted include changing camera angles, switching environments, and applying cinematic zooms while preserving the original scene context. Google showcased early developer use cases, including multi-perspective video generation from a single subject across roughly 20 different angles.
The launch expands Google's push into generative video as competition intensifies among major AI labs to offer multimodal tools for creators and developers.
