Loading...
Describe the image you want to create
0/20000Create an image from a written prompt, or add up to 14 references to guide an edit. Nano Banana 2.1 on Nana Banana AI offers 1K, 2K and 4K output, flexible aspect ratios, and a choice of JPG or PNG.
Start with words, then add visual references when a subject, setting or style needs more context.
Describe your subject, composition, lighting and intended changes in a prompt of up to 20,000 characters. A prompt is required for every generation.
Add up to 14 JPEG, PNG or WebP reference images, up to 30 MB each. Name each image's role in the prompt so the edit has a clear direction.
Select Auto or one of 14 fixed aspect ratios, then choose 1K, 2K or 4K resolution and JPG or PNG format before generating.
Nano Banana 2.1 is available alongside Nano Banana 2 in the model selector. This page opens the 2.1 model. New settings start at 1K resolution, Auto aspect ratio and JPG output; you can adjust each option before submitting.
Use reference images to show the subject to retain, a setting to borrow or a color palette to explore. Describe what should stay and what should change, then review the generated result against your inputs.
The 20,000-character prompt limit leaves room for composition, material, color and editing requirements.
The generation price depends on output resolution. Adding reference images does not add a separate charge.
Match each request to the image you need and check the displayed credit cost.
On Nana Banana AI, 1K costs 2 credits per image, 2K costs 4, and 4K costs 6.
Use a fixed ratio for a known placement, including 1:4, 4:1, 1:8 and 8:1 layouts. Auto is available when you want the model to choose the ratio.
Select the output format in the generator and download the completed image in that format. PNG output does not by itself request a transparent background.
Adapt these written examples to your own project and inspect the result before using it.
Choose Text to Image, 1K and 16:9. Prompt: “A small yellow bicycle leaning against a white wall, soft morning shadows, a potted plant on the left, spacious composition with empty space on the right, editorial illustration.”
Upload your portrait, select Image to Image, and write: “Replace the jacket with a yellow raincoat. Preserve the person's identity, pose, framing and lighting.” Choose the resolution and format, then review the face, garment and background in the result.
Prepare a brief, set the output and review the generated image.
Select Text to Image or Image to Image. Describe your image or edit, and upload up to 14 references if needed.
Pick the aspect ratio, 1K/2K/4K resolution and JPG/PNG format. Confirm the required credits before generating.
Sign in, submit your request and wait for completion. Check the result and download it; the task is also available in My Creations.
Input limits, supported output settings and generation credits.
Nana Banana AI charges 2 credits for a 1K image, 4 for 2K and 6 for 4K. Reference images do not add a separate charge.
You can upload up to 14 JPEG, PNG or WebP images, each up to 30 MB. Public HTTP or HTTPS image URLs can also be used; the source must be accessible and meet the same format and size limits.
There are 15 choices including Auto: auto, 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 and 21:9. Resolution choices are 1K, 2K and 4K.
Yes. Both text generation and image editing require a nonempty prompt, with a limit of 20,000 characters. Explain what to preserve and what to change in the references.
New Nano Banana 2.1 settings use 1K resolution, Auto aspect ratio and JPG output. PNG is also available. This model does not offer a Google Search setting.
A confirmed failed task receives a refund of its generation credits. A slow response or a temporarily unavailable status check does not by itself mean the task failed; you can check the existing task again.
Start from text or reference images, choose the output settings, and generate with Nano Banana 2.1.
