ChatGPT
Perplexity
Claude
Gemini
Nano Banana 2 Lite is Google's fastest, most affordable image model (gemini-3.1-flash-lite-image), but where should you actually run it?
There are three solid paths: Google's own AI apps, a developer API, and the OpenArt all-in-one creative platform. Here's how to use each.
Key takeaways
- Nano Banana 2 Lite is Google's fastest, cheapest image model (gemini-3.1-flash-lite-image).
- Easiest start: the Gemini app's Flash-Lite mode or Google AI Studio, no code needed.
- Developers can call it through the Gemini API for about $0.034 per 1K image.
- OpenArt is the best all-in-one home, with many top models under a single plan.
- OpenArt also adds 4 simultaneous generations per model, editing, and upscaling.
Option 1: Google's Gemini app and AI Studio
The quickest way to try Nano Banana 2 Lite is through Google's own apps.
In the Gemini app, switch to Flash-Lite mode, type a prompt, and generate, with no setup required. It's the easiest route for casual, everyday image creation on web or mobile, and a great way to see what the model can do before committing to anything.
For a bit more control, open Google AI Studio, start a new prompt, and select the gemini-3.1-flash-lite-image model. AI Studio is the best place to experiment with prompts, tune your inputs, and preview results before you build anything, and you can copy the generated code straight into your own project. Both are free to experiment with, which makes them the natural first stop.
Option 2: Plug into the Gemini API (technical setup)
To generate images programmatically, whether for an app, a batch job, or an automated pipeline, call Nano Banana 2 Lite through the Gemini API.
First, grab an API key in Google AI Studio and install the SDK:
pip install google-genai
Then generate an image with a few lines of Python:
from google import genai
client = genai.Client(api_key="YOUR_API_KEY")
response = client.models.generate_content(
model="gemini-3.1-flash-lite-image",
contents="A serene mountain lake at sunrise, soft morning light",
)
for part in response.candidates[0].content.parts:
if part.inline_data:
with open("output.png", "wb") as f:
f.write(part.inline_data.data)
At roughly $0.034 per 1K image, it's cost-efficient for high-volume workloads. Check Google's Gemini API image-generation docs for the latest parameters and limits. This route is ideal for product features, bulk content generation, and any workflow that needs images created on the fly.
Option 3: Use it on OpenArt (the best all-in-one option)
Google's tools are great for running Nano Banana 2 Lite on its own. But most creators don't stick to a single model, and that's where OpenArt stands out as the most productive place to use it.
Generating an image on OpenArt takes three simple steps:
Choose the model.
Open OpenArt's image generator and select Nano Banana 2 Lite from the model menu, then pick your format: vertical for Reels and Shorts, square for feed posts, or widescreen for thumbnails and banners.
Write your prompt.
Describe your subject, style, lighting, background, mood, and any in-image text you want, and upload reference images to fuse a character, product, or style into the shot.
Generate and refine.
Generate in near real time, then iterate in tight loops or open Edit Image for precise local changes, like swapping a background or fixing a label, before you download.
OpenArt also gives you more than a standalone generator:
Many models, one plan:
Alongside Nano Banana 2 Lite, OpenArt gives you Nano Banana Pro, GPT Image 2, Recraft V4, and more under a single subscription, with no separate accounts or API bills to juggle. Prototype on Lite, then switch to a heavier model when a project needs it.
Queue 4 generations at once, per model:
Run up to four generations simultaneously for each model, so you explore variations in parallel instead of one at a time, a perfect match for Lite's fast, iterative style.
Advanced editing and upscaling:
Refine results with built-in post-generation editing tools, and upscale beyond Lite's native 1K output for crisp, high-resolution exports, the finishing step Google's raw model doesn't include.
Put together, that means you get Lite's speed and low cost plus everything you need to take an image from rough draft to final export, without leaving the platform.
Start using Nano Banana 2 Lite today
If you just want to dabble, the Gemini app is the fastest start; if you're building software, the Gemini API is the way in. But if you create regularly and want every top model, parallel generations, and pro editing in one place, OpenArt's AI image generator is the best home for Nano Banana 2 Lite.