Nano Banana 2 Lite is a fast, low-cost AI image model from Google that turns a text prompt into a finished picture in about four seconds for roughly $0.034 per image. It is the speed tier of the Nano Banana family, built for rapid ideation, prototyping, and high-volume work — and you can try it for free in the generator below, no card required.
Launched June 30, 2026 · Model ID: gemini-3.1-flash-lite-image
Speed and cost, without throwing away the things that make Nano Banana useful.
The Lite model returns a text-to-image result in about four seconds, so you can iterate on a concept as fast as you can retype the prompt. That speed is the whole point of the family's lightest tier: it removes the wait between an idea and a draft, which makes it ideal for brainstorming, mood boards, and any workflow where you generate dozens of variations in a single session.
At roughly three cents per 1K-resolution image, this is the cheapest model in the Nano Banana lineup. The low per-image price is what makes high-volume pipelines, automated content, and large prototyping batches affordable — when you are generating thousands of images, the gap between a premium model and a lightweight one decides whether the project is viable at all.
Even tuned for throughput, the model keeps reliable prompt adherence, character consistency across continuous image streams, and legible in-image text with localized typography. Google also gave it upgraded world knowledge for drafting rough data visualizations and contextual layouts, so the output stays useful for real work instead of falling apart the moment you ask for text or a consistent character.
From a blank prompt to a finished image in seconds. There is no setup and no learning curve — pick the model, type, and generate.
Describe the image you want in plain language. The model has strong prompt adherence, so you can be specific about subject, style, composition, and any text you want rendered inside the image.
Select the Lite model in the generator and hit create. The result comes back in about four seconds, so you can compare several directions in the time a heavier model takes for one.
Because image editing is built into the same model, you can refine the result, adjust details, and keep characters consistent across a series — all without switching tools or paying a premium price per attempt.
Three models, one family. Lite is built for speed and cost, Nano Banana 2 for everyday balance, and Pro for the highest-quality professional work. Here is how to pick the right one for the job.
| Feature | Nano Banana 2 Lite | Nano Banana 2 | Nano Banana Pro |
|---|---|---|---|
| Best for | Speed, cost, high volume | Everyday balanced use | Complex professional work |
| Text-to-image speed | ~4 seconds | Fast | Quality-first |
| Cost per 1K image | $0.034 (lowest) | Mid-range | Premium |
| Resolution focus | Optimized for 1K | Balanced | Up to 4K |
| Text-to-image | |||
| Image editing | |||
| Character consistency | |||
| Legible in-image text | |||
| Gemini API access | |||
| Replaces original Nano Banana |
Specs reflect Google's launch details for the Lite model (gemini-3.1-flash-lite-image) on June 30, 2026. Nano Banana 2 and Pro positioning follow Google's model guidance.
Google's newest and leanest image model — the speed tier of the Nano Banana family, launched June 30, 2026.
The public name maps to gemini-3.1-flash-lite-image, the fastest and most cost-efficient model in Google's Nano Banana lineup. It runs on the Gemini 3.1 Flash Lite architecture and is served through the same Gemini API family as the rest of the range.
It targets rapid ideation, interactive prototyping, visual drafting, and high-throughput workflows where speed and cost matter more than maximum fidelity. This is the model you reach for when you need many images quickly, not a single hero shot.
Beyond text-to-image, the model handles image editing, so you can generate and refine inside the same fast, low-cost tool. There is no separate editing step and no premium surcharge for iterating on an existing image.
Google positions it as the recommended replacement for the original Nano Banana (gemini-2.5-flash-image), which is now the legacy model. If you were building on the first Nano Banana, this is the natural upgrade path.
It is tuned for jobs where speed and cost are the real constraints. These are the workflows where it shines.
Generate a wide spread of visual directions in minutes. Because each image costs about three cents and arrives in seconds, you can explore far more concepts before committing to one.
Drop placeholder visuals into product mockups, decks, and design drafts on the fly. The model keeps layouts contextual and text legible, so prototypes read clearly.
Produce large batches of images for catalogs, listings, ads, and automated pipelines where per-image cost decides whether the project is even feasible.
Spin up campaign variations, social posts, and thumbnail options quickly, then hand the winners to a higher-fidelity model only when you need the final render.
Keep characters consistent across a sequence of frames thanks to the model's character consistency, which makes it practical for storyboards and early concept art.
Use the upgraded world knowledge to draft rough charts, diagrams, and contextual layouts when you need a quick visual stand-in rather than a polished, final asset.
The launch facts that define it.
Typical time to generate a text-to-image result.
Cost per 1K-resolution image — the lowest price in the Nano Banana family.
Modes built in: text-to-image generation and image editing.
When Google launched the model (gemini-3.1-flash-lite-image).
Common questions about Google's fastest, cheapest Nano Banana image model.
Generate your first images in seconds — free starter credits, no card required.
We use cookies to improve your experience on our website. By browsing this website, you agree to our use of cookies. Learn more