Google unleashed Gemini Omni, its new AI video generation tool, with the June Pixel Drop. Billed as 'like Nano Banana for video,' Omni enables users to create video content through conversational prompts, supporting text, image, and audio inputs. Early users report a polished experience that surpasses existing models like Veo.

Gemini Omni Flash is the specific model powering this video generation within the Gemini app, also appearing in Google Flow and YouTube Shorts. While currently video-focused, future iterations aim for broader creative outputs. Availability is tied to Google AI subscription tiers, with a free version offering limited use, but it's accessible on any smartphone, not just Pixel devices.

The tool supports multiple input types, allowing users to upload photos or videos, record audio snippets, and provide text prompts. This multi-modal approach simplifies the creation process. After generation, videos can be shared or downloaded directly from the app, making AI video creation more accessible for everyday users.