Nano Banana 2: AI Image Generator & Editor Online
Bottom line
Text- and image-to-image with up to 14 reference images, auto framing, and aspect ratios from 1:8 to 8:1. Generate from text or edit an existing image with Nano Banana 2 in one place — VdoBloom shows the credit cost for either mode before you confirm.
Most image generators stop at 21:9. Nano Banana 2 keeps going — to 4:1 and 8:1 horizontal strips and their 1:4 and 1:8 vertical counterparts — which puts web leaderboards, skyscraper ad units, panoramic headers, and side-rail graphics inside a single generation instead of an outpainting exercise. Fifteen aspect options in total, including an auto setting that lets the model choose framing from your prompt and inputs, each renderable at 1K, 2K, or 4K.
The other headline number is fourteen: the count of reference images a single job accepts. That turns Nano Banana 2 into a compositing tool as much as a generator — character from one image, wardrobe from another, product from a third, environment from a fourth, merged into one coherent frame. It handles text-to-image and image-to-image in the same interface, so the line between generating and editing mostly disappears.
It rates 4.3 with VdoBloom users and carries strict provider content filters, so restricted material gets refused in any mode. Credit cost for each configuration is shown before the run starts.
What creators use Nano Banana 2 for
- Website skyscraper banners and leaderboard strips at 8:1 or 1:8 without stitching tiles together.
- Compositing a character, outfit, and setting from up to 14 reference images into one coherent scene.
- Letting the auto aspect setting pick output framing during image-to-image edits of an uploaded source.
How to use Nano Banana 2 on VdoBloom
- 1Open the Text to Image tool and select Nano Banana 2.
- 2Write your prompt (or upload an image to edit).
- 3Set resolution (1K/2K/4K), aspect ratio, review the credit cost, and generate.
Frequently asked questions
How many reference images can Nano Banana 2 take?
Up to 14 in a single job, which is what makes it suited to compositing — pulling a character from one image, wardrobe from another, and a setting from a third, then merging them into one frame.
What does the "auto" aspect ratio setting do?
Instead of locking a fixed ratio, auto lets the model choose the output framing based on your prompt and any input images. It is most useful in image-to-image work where you want framing driven by the source material.
Why would I need 8:1 or 1:8 aspect ratios?
Web leaderboards, skyscraper ad units, site headers, and panorama strips all live in those extreme formats. Nano Banana 2 renders them natively at 1K, 2K, or 4K instead of requiring outpainting or tile stitching.
How does Nano Banana 2 handle sensitive content?
Strictly. It carries the CENSORED label, meaning strict provider filters, and refusals on restricted material apply across both text-to-image and image-to-image modes.