Google Builds Nano Banana 2 Into Earth to Reimagine Any Location
Google Earth now lets you generate AI images of any location on Earth using Nano Banana 2, grounded in real satellite and 3D data.
- Google has integrated Nano Banana 2 image generation into Google Earth on the web, live globally and free today.
- Users zoom into any location, tap "create image," and type a prompt; the model generates images grounded in real satellite and 3D data.
- Nano Banana 2 (Gemini 3.1 Flash Image) combines Pro-level quality with Flash speed, supporting up to 4K resolution and 5-character consistency.
- Use cases include historical visualization, real estate rendering, urban planning, location infographics, and creative reimagining.
- The feature is currently web-only, excluding the Google Earth mobile app despite 500M+ Play Store downloads.
- All generated images are watermarked with SynthID and paired with C2PA Content Credentials for AI provenance tracking.
Google just made Google Earth a lot more interesting. The company has integrated Nano Banana 2 image generation directly into Google Earth on the web, letting anyone zoom into any spot on the globe, type a prompt, and get a photorealistic AI-generated image grounded in actual satellite, aerial, and 3D map data. It is live globally right now, for free.
What just happened
For the first time, you can generate custom images using Google Earth's satellite, aerial, and 3D imagery alongside Nano Banana, which creates concepts grounded in the real world. The workflow is as simple as it sounds: zoom in to a place in Google Earth on web, tap "create image," and type whatever you want to see.
The key word here is grounded. This is not a generic text-to-image tool. The model uses the actual geometry and visual context of the location you are looking at as a base, then layers your prompt on top. That means a prompt like "add a modern lakefront cabin" does not produce a generic cabin floating in a void. It produces a rendering of that specific lot, at that specific angle, with the surrounding landscape intact.
The engine under the hood
Google DeepMind built Nano Banana 2 as a model that combines the advanced features of Nano Banana Pro with the speed of Gemini Flash. More precisely, Nano Banana 2 is more accurately known as Gemini 3.1 Flash Image; the previous Nano Banana models were based on the 3.0 branch.
The architectural difference from traditional diffusion models matters here. Built on Google's Gemini 3.1 Flash Image foundation, Nano Banana 2 reasons about composition, lighting, and spatial relationships before rendering. Google's Gemini 3.1 Flash Image architecture generates vibrant, high-fidelity visuals at speed, combining multimodal reasoning with Flash-optimized inference. In plain terms: instead of treating your prompt as a bag of weighted keywords the way older diffusion models do, it reads your prompt the way a language model would, understanding intent and context before it draws anything.