Google’s Nano Banana 2 Lite Now Makes AI Images in Just Four Seconds
Google has added two new generative AI models to its lineup, and they’re built for speed. Nano Banana 2 Lite generates images in four seconds and costs $0.034 per image at 1K resolution. Alongside it, the company also launched Gemini Omni Flash, which brings video generation and editing into the API for the first time.
Nano Banana 2 Lite is designed for quick brainstorming and high volume developer workflows. It runs through the API under the name gemini-3.1-flash-lite-image. Despite the focus on speed, Google says the model still delivers reliable prompt following, consistent character rendering, and readable text in generated images. The model is rolling out beyond developer tools too, reaching AI Mode in Google Search, the Gemini app, NotebookLM, Google Photos, Stitch, Google Flow, and Google Ads.
With this release, the Nano Banana family now has three models in production. Nano Banana 2 sits in the middle as the balanced option for quality and cost, while Nano Banana Pro is aimed at more complex, professional jobs requiring stronger control. Nano Banana 2 Lite now fills the budget and speed slot, replacing the original Nano Banana model.
The second release, Gemini Omni Flash, was first previewed at Google I/O and is now live in the Gemini API and Google AI Studio. Gemini Omni Flash lets developers generate and edit videos up to ten seconds long through text prompts via the API for $0.10 per second of output, a price that matches Google’s Veo 3.1 Fast. The model can mix text, image, and video inputs, and it taps into Gemini’s broader knowledge base during generation. Google notes some current limits too, including no audio references yet, no scene extensions, and inconsistent handling of video reference clips.
Where things get interesting is when the two models work together. Developers can quickly generate images with Nano Banana 2 Lite and pass them as references to Gemini Omni Flash, which then animates them into video. Google has even built demo apps to show this off, including one that places users at famous landmarks from a selfie, another that turns room photos into animated interior design concepts, and a third that converts product photos into short e-commerce videos.
Both models carry SynthID watermarks to flag AI generated content, and they’re available now in Google AI Studio, the Gemini API, and the Gemini Enterprise Agent Platform.
