Google Introduces New AI Models for Faster Image and Video Creation
Google has introduced two new generative AI models designed to make image and video creation faster and more affordable for developers and consumers. The launch expands the company’s AI portfolio with separate models for image generation and video production.
Nano Banana 2 Lite is now generally available as Google’s fastest image generation model. It can create a 1K-resolution image in as little as four seconds and costs $0.034 per image. Meanwhile, Gemini Omni Flash has entered public preview, offering video generation and conversational editing for $0.10 per second of video output.
New features improve content creation
Nano Banana 2 Lite, officially named Gemini 3.1 Flash-Lite Image, replaces the original Nano Banana model. Additionally, it delivers improved visual quality, better character consistency and more accurate text rendering. Google designed the model for high-volume workloads, including advertising experiments and large-scale social media applications.
Meanwhile, Gemini Omni Flash enables users to generate and edit videos using natural language prompts. The model combines text, image and video inputs while producing clips up to 10 seconds long. Furthermore, it generates audio alongside video and maintains consistent characters and visual styles throughout edits. Google also plans to support longer video durations in future updates.
Models roll out across Google products
Both AI models are available through Google AI Studio, the Gemini API and the Gemini Enterprise Agent Platform. In addition, Google is rolling them out across the Gemini app, AI Mode in Search, NotebookLM, Google Photos, Google Flow and Google Ads.
NotebookLM now uses Nano Banana 2 Lite to power its new Short Video Overviews feature. The tool converts uploaded documents into approximately 60-second vertical videos with narration and educational animations. Initially, the feature is available to English-speaking users aged 18 and older on the web, Android and iOS.
Google also confirmed that companies including Adobe, WPP and Figma have started integrating the new models into their products. Moreover, both models include SynthID watermarks and C2PA content credentials by default to improve transparency for AI-generated content.
