Nano Banana 2.1
nano-banana-2-1

Nano Banana 2.1 is Google’s high-efficiency AI image model for image generation and editing, with improved visual quality, prompt adherence, character consistency, text rendering, multi-reference support, and up to 4K output.

Pricing: 4 credits ($0.02) for 1K, 6 credits ($0.03) for 2K, and 9 credits ($0.045) for 4K — 1K/2K 40% and 4K 60% below official pricing. High-tier top-ups come with bonus credits that lower the effective price by about 10%.
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)

Resolution of the generated image. Higher resolutions take longer to generate.

Format of the output image

Output
output typeimage
Generated output
Examples

Explore different use cases and parameter configurations

README

Complete guide to using nano-banana-2-1

Nano Banana 2.1 API for Text-to-Image and Image Editing

Original image
ParameterDetails
modelRequired. Set to nano-banana-2-1.
promptRequired. Text prompt up to 20000 characters.
image_inputOptional. Up to 10 input images to transform or use as reference. Pass file URLs after upload (not file content) in image/jpeg, image/png or image/webp format, up to 30.0MB each. Without image_input the request is text-to-image; with it, the same model edits or combines the input images.
aspect_ratioOptional. 1:1, 2:3, 3:2, 1:4, 4:1, 3:4, 4:3, 4:5, 5:4, 1:8, 8:1, 9:16, 16:9, 21:9 or auto. Default: auto.
resolutionOptional. 1K, 2K or 4K. Default: 1K.
output_formatOptional. png or jpg. Default: jpg.
4K Panorama at 4:1
Infographic with Exact Labels
Two Input Images Combined
Chalkboard Text Recolor
Nano Banana 2.1Nano Banana 2Nano Banana Pro
Model IDnano-banana-2-1nano-banana-2nano-banana-pro
Input imagesUp to 10 (optional)Up to 10 (optional)Up to 8 (optional)
Aspect ratios1:1, 2:3, 3:2, 1:4, 4:1, 3:4, 4:3, 4:5, 5:4, 1:8, 8:1, 9:16, 16:9, 21:9, auto1:1, 2:3, 3:2, 1:4, 4:1, 3:4, 4:3, 4:5, 5:4, 1:8, 8:1, 9:16, 16:9, 21:9, auto1:1, 2:3, 3:2, 3:4, 4:3, 4:5, 5:4, 9:16, 16:9, 21:9, auto (no 1:4, 4:1, 1:8, 8:1)
Default aspect ratioautoauto1:1
Resolution1K, 2K, 4K (default 1K)1K, 2K, 4K (default 1K)1K, 2K, 4K (default 1K)
Output formatpng, jpg (default jpg)png, jpg (default jpg)png, jpg (doc says to set explicitly)

You can test it for free: new Kie.ai users receive 80 free credits for testing. After that, generations draw on credits you top up. You can try this model in the playground on this page without writing code. Kie.ai provides this model as an API with a playground; there is no app to download or install.

The price for Nano Banana 2.1 is shown in the pricing card above. Your account has one credit balance that all generation services draw from; check it with the Get Remaining Credits API.

Google describes Nano Banana 2.1 as "the more efficient counterpart to Gemini 3 Pro Image (Nano Banana Pro)." On Kie.ai, nano-banana-2-1 accepts up to 10 input images, includes the 1:4, 4:1, 1:8 and 8:1 aspect ratios, and defaults aspect_ratio to auto. nano-banana-pro accepts up to 8 input images, does not include 1:4, 4:1, 1:8 or 8:1, and defaults aspect_ratio to 1:1. Google's October 2026 model card reports a side-by-side human evaluation of text-to-image overall preference. The Elo scores are 1050 for Nano Banana 2.1 with thinking, 1015 for Nano Banana 2.1 without thinking, and 935 for Nano Banana Pro. Our sources do not state which thinking setting Kie.ai's endpoint uses. For cost on Kie.ai, see the pricing card for each model.

On Kie.ai, nano-banana-2-1 and nano-banana-2 have the same documented parameters, as shown in the comparison table above. Both accept up to 10 input images and the same 14 aspect ratios plus auto (default auto). Both offer 1K, 2K or 4K resolution (default 1K) and png or jpg output (default jpg). According to Google, Nano Banana 2.1 (based on Gemini 3.6 Flash) is an update to Nano Banana 2 (Gemini 3.1 Flash Image). Google lists improved visual quality and realism at 1K, 2K and 4K, as well as better prompt adherence, multi-turn character consistency, text rendering and infographic layout accuracy. Google also says it fixed tiling artifacts on wide and panoramic aspect ratios (1:4, 4:1, 1:8, 8:1) at 2K and 4K. Prices for both models are in their pricing cards.

Google describes Gemini Nano Banana 2.1 (gemini-nano-banana-2.1) as its latest high-efficiency image generation and conversational editing model and an update to Nano Banana 2. This comes from Google's documentation, last updated October 6, 2026. Google DeepMind's October 2026 model card adds that it is a member of the Gemini 3 series, based on Gemini 3.6 Flash.

Create a key on your Kie.ai API Key Management page. All models in the Kie.ai market use the same authentication method: an Authorization: Bearer YOUR_API_KEY header with the key from your Kie.ai API Key Management page. Send requests to POST /api/v1/jobs/createTask with the parameters listed in the parameter table above, and receive results via callBackUrl (recommended for production) or Get Task Details.

As shown in the parameter table above, the image_input parameter on Kie.ai accepts up to 10 input images as file URLs after upload (in image/jpeg, image/png or image/webp format, up to 30.0MB each). Google's own API documentation lists multi-image fusion with up to 14 reference images, but the Kie.ai limit is 10.

Yes. As listed in the parameter table above, the resolution parameter accepts 1K, 2K or 4K (default 1K), and the aspect_ratio parameter accepts 14 aspect ratios including 1:4, 4:1, 1:8 and 8:1, plus auto (default auto).

No. The Kie.ai API for this model has no mask parameter; see the parameter table above for the supported fields. Google's model card also lists partial instruction following in mask- or doodle-based editing as a known limitation.