Google Launches Nano Banana 2 Lite and Gemini Omni Flash

Google is expanding its generative media capabilities with the release of Nano Banana 2 Lite and Gemini Omni Flash. These two models are designed to help developers experiment, refine, and scale multimedia projects by connecting rapid image generation with advanced video creation and conversational editing. Both models are available now in Google AI Studio, the Gemini API, and the Gemini Enterprise Agent Platform.

Introducing Nano Banana 2 Lite

Nano Banana 2 Lite is the fastest and most cost-efficient image model in the Nano Banana family, built specifically for high-velocity developer pipelines and rapid ideation. It delivers text-to-image outputs in approximately 4 seconds and is priced at $0.034 per 1,000 images. Despite its focus on speed and budget-friendly performance, the model maintains reliable prompt adherence, consistent character representation, and legible in-image text rendering.
This model is intended to replace the first version of Nano Banana (gemini-2.5-flash-image). Beyond developer platforms, Nano Banana 2 Lite is rolling out to various Google consumer surfaces, including AI Mode in Search, the Gemini app, NotebookLM, Google Photos, Stitch, Google Flow, and Google Ads.

Gemini Omni Flash for Video Generation

Gemini Omni Flash brings multimodal reasoning to video generation and conversational editing. The model allows users to refine videos using natural language and supports inputs from text, images, and video to maintain scene consistency. It is priced at $0.10 per second of video output. While the model is currently in public preview, it offers capabilities such as text and action synchronization and the ability to leverage Gemini’s internal knowledge of history, biology, and narrative logic to construct videos.
Current limitations for Gemini Omni Flash include a 10-second limit on video generations and a lack of support for audio references or scene extensions within the Gemini API. While the API schema accepts video references up to 3 seconds, these are not yet correctly processed by the model.

Building End-to-End Multimedia Experiences

Developers can create comprehensive workflows by chaining these models together, such as using Nano Banana 2 Lite to generate an image and then passing that image to Gemini Omni Flash to animate it. By utilizing the Interactions API, developers can maintain session history and context, allowing for up to three sequential edits in a single workflow.
Google has provided several demo applications to showcase these capabilities. The "Anywhere" app uses Nano Banana 2 Lite to place users in iconic locations, while "Space Lift" allows for interior design reimagining. Additionally, the "Omni product studio" demo illustrates how static images can be converted into cinematic e-commerce videos. To ensure transparency, both models are built on Google’s secure infrastructure and utilize SynthID watermarking, allowing users to verify AI-generated content across various platforms.