Nano Banana 2 Lite
nano-banana-2-lite

Nano Banana 2 Lite, aka Gemini 3.1 Flash-Lite Image, is a newly released image model from Google DeepMind built for fast 1K image generation, efficient creative iteration, prompt-based image editing, and responsive visual output in product-facing creative scenarios.

Pricing: Nano Banana 2 Lite — now just 4 credits ($0.02) for 1K, about 59.5% of the Google official price, saving roughly 40.5%. High-tier top-ups (+10% bonus) bring the effective pricing down to ~$0.018 for 1K.
Input

A text description of the image you want to generate

0/20000

Click to upload or drag and drop

Supported formats: JPEG, PNG, WEBP Maximum file size: 30MB; Maximum files: 10

Input images to transform or use as reference (supports up to 10 images)

Aspect ratio of the generated image

Output
output typeimage
Generated output
Examples

Explore different use cases and parameter configurations

README

Complete guide to using nano-banana-2-lite

Nano Banana 2 Lite API: Text-to-Image and Image Editing

Original image
ParameterValue
Modelnano-banana-2-lite (Google model ID: gemini-3.1-flash-lite-image)
EndpointPOST /api/v1/jobs/createTask, then Get Task Details or a callBackUrl for the result
PromptRequired; up to 20,000 characters
Reference images (image_urls)Optional; up to 10 image URLs (hosted files, not raw file content); JPEG, PNG or WebP; up to 30 MB each. Omit or leave empty for text-to-image
Aspect ratioRequired; 1:1, 1:4, 1:8, 2:3, 3:2, 3:4, 4:1, 4:3, 4:5, 5:4, 8:1, 9:16, 16:9, 21:9, or auto to let the model decide
OutputOne JPEG image per task; the format cannot be set
Output sizeAbout 1 megapixel (1K); there is no resolution parameter. In our tests on Kie.ai: 1024×1024 at 1:1, 1200×896 at 4:3, 1376×768 at 16:9, 2064×512 at 4:1
WatermarkGoogle applies SynthID watermarking to Nano Banana 2 Lite images
PricingShown in the pricing card above
Legible In-Image Text: A Café Chalkboard Menu
A Character From a Short Prompt: An Orange Robot Mascot
Character Consistency With a Reference Image: The Same Robot in a Library
Wide Formats: A Panoramic Website Banner With a Headline
Nano Banana 2 LiteNano Banana 2Nano Banana
Google model IDgemini-3.1-flash-lite-imagegemini-3.1-flash-imagegemini-2.5-flash-image
Google's positioningHigh-volume workflowsThe generalist workhorseLegacy model; Google recommends upgrading to Nano Banana 2 Lite
LMArena text-to-image score (Oct 7, 2026)1250Not listed without web search1150
LMArena single-image edit score (Oct 6, 2026)1314Not listed without web search1293

Nano Banana 2 Lite costs 4 credits (about $0.02) per image. Every image comes out at about 1 megapixel (1K), and text-to-image and editing cost the same. Prices may be updated; the pricing card at the top of this page always shows the current rate.

Nano Banana 2 Lite is Google's image model for text-to-image generation and image editing, released on June 30, 2026 with the model ID gemini-3.1-flash-lite-image. It is the Nano Banana model Google built for high-volume work; capabilities, parameters and examples are covered above.

Google positions Nano Banana 2 as its generalist workhorse and Nano Banana 2 Lite for high-volume workflows. Leaderboard scores for both versions, where listed, are in the comparison table above; compare both on your own prompts before switching.

Not on Kie.ai: each generation uses credits, as shown in the pricing card above. New Kie.ai users receive 80 free credits for testing.

No. Nano Banana 2 Lite is a proprietary Google model: as of October 2026, Google offers it through its own apps and APIs and has not released model weights, so there is no official download. Kie.ai provides this model as an API with a playground; there is no app to download or install.

Google offers it in Google AI Studio, the Gemini API and Gemini Enterprise Agent Platform, and inside products such as the Gemini app and Google Photos. On Kie.ai you call it as the nano-banana-2-lite API model. You can try this model in the playground on this page without writing code.

Create an API key, then send a createTask request with the model nano-banana-2-lite, a prompt and an aspect_ratio. The steps above show the full flow, and the parameter table lists every input.

Sign up on Kie.ai and create a key on the API Keys page (kie.ai/api-key), then send it in every request as the header Authorization: Bearer YOUR_API_KEY. This is a Kie.ai key, not a Google key: one key and one credit balance work for every model on Kie.ai. The full reference is on the Nano Banana 2 Lite page in the Kie.ai docs. Requests are asynchronous: createTask returns a taskId, and you get the result through your callBackUrl or by polling Get Task Details.

Yes. Pass one or more reference images in image_urls together with an instruction; the same model handles text-to-image and editing. The parameter table above lists the input limits, and the library example shows a reference-image edit.

Images come out at about 1 megapixel (1K), and there is no resolution setting; the aspect ratio decides the width and height. The parameter table above lists the sizes we measured.