Google has expanded its generative AI portfolio with the launch of Nano Banana 2 Lite, its fastest and most cost-efficient Gemini image generation model, alongside Gemini Omni Flash, a new multimodal AI model for conversational video creation and editing.
The launches are aimed at developers and enterprises building AI-powered multimedia applications, while the new capabilities are also rolling out across Google's consumer products, including the Gemini app, AI Mode in Search, NotebookLM, Google Photos, Google Flow and Google Ads.
"Today, we're making it faster and easier to experiment, refine and scale your ideas with two major releases," Alisa Fortin, Product Manager at Google DeepMind, and Anish Nangia, Product Manager at Google DeepMind, said in a blog post on Tuesday.
ALSO READ: Meta Contractors Pose As Teens To Test ChatGPT, Gemini On Topics Of Suicide, Sex And Drugs
Nano Banana 2 Lite has been designed for high-volume image generation where speed and affordability are key priorities. According to Google, the model can generate text-to-image outputs in about four seconds and costs $0.034 per 1,000-resolution image, making it suitable for rapid prototyping, large-scale content creation and budget-conscious deployments.
Google said the model retains strong prompt adherence, character consistency and text rendering despite being optimised for speed. The company is recommending developers using the older Nano Banana (Gemini 2.5 Flash Image) model migrate to Nano Banana 2 Lite for improved quality, lower latency and reduced costs.
Alongside the image model, Google has also opened Gemini Omni Flash to developers through Google AI Studio, the Gemini API and the Gemini Enterprise Agent Platform after previewing it at Google I/O.
The model enables conversational video editing through natural language prompts and supports multimodal inputs, allowing users to combine text, images and video references while generating or editing videos. Google has priced Omni Flash at $0.10 per second of generated video, matching the pricing of Veo 3.1 Fast.
According to the company, the model can synchronise text and visual actions, incorporate real-world knowledge into generated videos and maintain scene consistency using multiple input formats.
ALSO READ: Gadkari Says Technology Can Help With Enhancing Road Safety In India
Google said Omni Flash currently supports video generation of up to 10 seconds. Longer video durations, audio reference uploads and enhanced scene consistency are among the features still under development.
The company also highlighted workflows that combine both models, allowing developers to generate images with Nano Banana 2 Lite before converting them into animated clips using Gemini Omni Flash. Demo applications showcased by Google include AI-powered travel visualisations, interior design concepts and e-commerce product videos.
Google added that both models include SynthID watermarking to improve transparency and help users identify AI-generated content.
Essential Business Intelligence, Sharp Market Insights, Practical Personal Finance Advice, Daily Fuel, Gold and Silver Prices and Latest Stories — On NDTV Profit.