Google has launched Nano Banana 2 Lite, a new AI image generation model that produces images in about four seconds for roughly $0.034 apiece, alongside Gemini Omni Flash, a companion model that brings AI video generation and editing to the Gemini API for the first time.
What Nano Banana 2 Lite does
Google built Nano Banana 2 Lite for fast ideation and high-throughput developer pipelines, where speed and cost matter more than squeezing out every last bit of image quality. Text-to-image generation takes about four seconds, and each 1K-resolution image costs $0.034 through the API, where the model is listed as gemini-3.1-flash-lite-image. Despite the speed focus, Google says the model still delivers reliable prompt following, consistent character rendering across multiple images, and legible text, three areas where cheaper, faster image models often stumble.
Nano Banana 2 Lite is also rolling out across Google's own consumer products, including AI Mode in Google Search, the Gemini app, NotebookLM, Google Photos, Stitch, Google Flow, and Google Ads, putting the same fast, low-cost model behind a wide range of everyday features.
Where Nano Banana 2 Lite fits in Google's AI image lineup
The new release brings Google's Nano Banana family to three production models, each aimed at a different use case:
- Nano Banana 2 Lite ($0.034 per 1K image), the fastest and cheapest option, built for high-volume generation.
- Nano Banana 2 ($0.067 per 1K image), Google's all-rounder, balancing quality and cost.
- Nano Banana Pro ($0.134 per 1K or 2K image), aimed at complex, professional work, with what Google calls the strongest control and most advanced reasoning.
Google now considers the original Nano Banana (Gemini 2.5 Flash Image) outdated. Developers can pick a model based on whether they're prioritizing speed, overall quality, or raw cost efficiency for a given project.
Gemini Omni Flash brings video generation to the API
Alongside the image model, Google is rolling out Gemini Omni Flash, first previewed at Google I/O and now available to developers through the Gemini API and Google AI Studio. The model merges Gemini's multimodal reasoning with the ability to generate and edit video from text prompts, priced at $0.10 per second of output, matching the rate for Veo 3.1 Fast.
Google highlights conversational video editing through natural language, the ability to mix input formats like text, images, and video in a single request, and access to Gemini's broader world knowledge as key strengths, with text and graphics able to sync directly with on-screen action. The model currently tops out at ten-second clips, and some features remain unfinished: audio references and scene extensions aren't supported in the API yet, and while the schema accepts short video references, Google says the model doesn't process them correctly yet. Character consistency across scene changes or camera moves is also still limited.
Chaining the two AI models together
Google is pitching the biggest value in using Nano Banana 2 Lite and Gemini Omni Flash together: generate a still image quickly and cheaply with the former, then feed it into the latter to animate it into video. The Interactions API, now Google's default interface for Gemini models and agents, preserves session history and context, allowing up to three consecutive edits within a single workflow.
To demonstrate the combo, Google released three sample apps. "Anywhere" places a user at a famous landmark using a selfie and then animates the scene. "Space Lift" turns room photos into interior design concepts and then into video walkthroughs. "Omni Product Studio" converts static product photos into short e-commerce videos. Developers experimenting with these workflows can find more coverage of the underlying models in our ongoing AI news section, and teams evaluating the API pricing for production use may also want to track broader shifts in AI business spending as generative tools get cheaper to run.
Both models embed SynthID watermarks to flag AI-generated content, which can be verified through the Gemini app, Gemini in Chrome, or Google Search. Nano Banana 2 Lite and Gemini Omni Flash are available now in Google AI Studio, the Gemini API, and the Gemini Enterprise Agent Platform.
Frequently asked questions
How much does Nano Banana 2 Lite cost per image?
Nano Banana 2 Lite costs $0.034 per image at 1K resolution, making it the cheapest model in Google's current Nano Banana lineup.
Can Gemini Omni Flash generate long videos?
Not yet. The model currently generates clips up to ten seconds long, and features like audio references and scene extensions aren't fully supported in the API.
