Google has unveiled Gemini Omni, a new artificial intelligence-powered video creation tool designed to generate and edit videos using text, images and voice prompts, as the company expands its Gemini AI ecosystem.

Announcing the tool on its official blog, Google said Gemini Omni is built to support “native multimodal video generation,” allowing users to create videos directly from simple instructions while combining text, visuals and audio inputs.

“Gemini Omni is our next step toward more capable multimodal generation,” Google said, describing the model as one designed to better understand and create across different media formats.

According to Google, the new model can generate short-form videos from written prompts, transform still images into moving visuals, and use voice input to shape scenes and storytelling.

The company furthered that the system is designed to better understand context across multiple formats, allowing users to create more detailed and coherent video outputs.

Google said Gemini Omni is aimed at creators, developers and businesses looking to integrate AI-generated video into products and workflows.

The company added that the tool could support use cases ranging from content production and marketing to education and software development, as demand grows for faster video generation tools.

The launch forms part of Google’s wider strategy to expand Gemini beyond chat and search into content generation, productivity and software development.

Industry analysts say the move places Google in direct competition with other technology companies building advanced AI video tools, as the race to commercialise generative AI continues.