Limitations:
Omni currently offers 10 second video generation, with longer duration video generation coming soon. Uploading audio references and scene extensions is not yet supported in the Gemini API for this model. Video references up to 3 seconds in duration are accepted by the API schema, but are not currently handled correctly by the model. There are some limitations in character consistency during scene changes and panning, and we are working to improve this.
Gemini Omni is available in public preview in Google AI Studio and Gemini API starting today. Check the developer documentation for a complete list of model features and region-specific limitations.
Build both models now
The real magic happens when you connect these models together. Use Nano Banana 2 Lite as a fast image generation model and pass the images as a reference to Gemini Omni Flash to animate them into high-quality videos. Additionally, using the Interaction API for these multi-turn experiences allows you to maintain session history and context, allowing users to stack up to three consecutive edits.
To help you get started, we’ve created some remixable demo apps that let you experience how you can combine both Nano Banana 2 Lite and Gemini Omni Flash into one workflow.
