Can I edit an image with Nano Banana on Kyeo AI?

Yes. A request with images uses `google/nano-banana-edit`; a request without images uses `google/nano-banana` to create a new image. Both start from the same model page.

How many images can I upload for an edit?

Start with 1 to 3 and treat 10 as an interface maximum.

When should I compare Nano Banana 2 or Pro?

Compare Kyeo's Nano Banana 2 and Pro when you need explicit 1K, 2K, or 4K tiers, search grounding, or simply want to assess a newer option. Test the same prompt and source material before choosing.

Nano Banana: 3-credit image generation and editing

AI image generator and editor

Google

Nano Banana

Nano Banana, also known as Gemini 2.5 Flash Image, provides text-to-image generation and natural-language editing on Kyeo AI. Requests without images use `google/nano-banana`; requests with images use `google/nano-banana-edit`. Both paths cost 3 credits and accept prompts up to 5,000 characters.

Capabilities

Text-to-Image and Image-to-Image

Credit cost

3 credits per request

Prompt limit

5,000 characters

Upload limit

Up to 10 images

Requests without images use `google/nano-banana`; requests with images use `google/nano-banana-edit`.

Text-to-image generation and editing each cost 3 credits per request.

Prompts allow 5,000 characters, with PNG or JPEG output and 11 aspect-ratio choices.

upstream interface accepts up to 10 editing images, but Google recommends no more than 3 inputs for best performance.

Try this model for free View credit pricing Back to model gallery

30-second overview

Nano Banana

3 credits per request

What it does best

Its defining tradeoff here is a fixed 3-credit cost rather than higher-resolution tiers or search grounding.

Best for

Use it to create an image from text, add, remove, replace, or restyle parts of an existing image, or test a composition with a small set of clearly assigned references. Character, product, and brand consistency must be checked rather than assumed.

At a glance

What is this model like?

This is a fixed-price generation and editing option with no resolution-priced tiers. Google now lists Gemini 2.5 Flash Image as a legacy model in its Gemini API guidance and advises direct API customers to move to a newer generation. Kyeo's route remains available, so it should be chosen with that lifecycle status and the actual test results in view.

What it does best

Its defining tradeoff here is a fixed 3-credit cost rather than higher-resolution tiers or search grounding.

Best for

Why use it on Kyeo AI

The form exposes the current boundaries: a 5,000-character prompt, PNG or JPEG output, 11 aspect ratios, and up to 10 images on the editing route. Kyeo selects the matching generation or editing path automatically.

Key facts

Quickly assess whether this model matches your workflow.

Also known as

The same model may appear under different names across documentation and community discussions; this list keeps them easy to search and compare.

google/nano-banana

Gemini 2.5 Flash Image

Google Nano Banana

Common search intents

These questions and keywords combine public references with common creative use cases to help you choose faster.

How do I run Nano Banana on Kyeo AI

What does a Nano Banana request cost on Kyeo AI

How does Nano Banana generate an image without an upload

What happens when I attach an image to Nano Banana

Do generation and editing have the same Nano Banana price

How many images does the Nano Banana editing route accept

When does Nano Banana Pro make more sense than the fixed-price model

How can I compare Nano Banana and GPT Image 1.5 editing results

What should I inspect after editing a product image with Nano Banana

Is Nano Banana the same as Gemini 2.5 Flash Image

Selection guide

Use these decision points when choosing a model.

Choose it when a fixed request cost matters

If the brief does not require a 1K, 2K, or 4K tier or search grounding, a 3-credit generation or targeted edit is a low-commitment way to test the result before continuing.

Compare newer options for resolution or search

Kyeo's Nano Banana 2 and Pro are the relevant comparisons when you need explicit 1K, 2K, or 4K output tiers or search grounding. Use the same source image with GPT Image 1.5 to compare a different vendor's editing behavior.

Start with 1 to 3 references

For multi-image edits, assign a clear role to each input and begin with a small set. Add more only when the extra material is necessary and the result remains controllable; a 10-image limit does not mean 10 images will be more reliable.

Popular comparisons

Common comparison questions help you find the practical differences quickly.

How does 3-credit Nano Banana compare with resolution-priced Pro

Should I use Pro when a Nano Banana job needs a 4K tier

Which projects are better suited to Nano Banana 2

Does a search-grounded brief call for Nano Banana 2 instead

How should I test Google and OpenAI image editors fairly

How do Nano Banana and GPT Image 1.5 input limits compare

Model comparison

Compare the current model with alternatives at a glance.

Dimension

Nano Banana

Nano Banana Pro

GPT Image 1.5

Current role on Kyeo

Fixed-price image generation and editing

Resolution-priced option with a 4K tier

OpenAI image generation and editing option

Workflow structure

One page switches between generation and editing based on image input

One model entry accepts either text or image input

One page selects the matching path for each input type

Current Kyeo price

3 credits for either generation or editing

8 credits at 1K or 2K; 14 credits at 4K

5 credits at 1K; 8 at 2K; 12 at 4K

Editing image count

Google recommends no more than 3 for best performance

Currently accepts up to 8 input images

Currently accepts up to 16 reference images

Best-fit requirement

Fixed-price generation or targeted edits without a resolution tier

Work that needs an explicit 1K, 2K, or 4K tier

Work that needs more resolution choices or search grounding

Practical usage insights

Practical observations based on real creative workflows.

An interface maximum is not a recommended input count

Google's guidance for Gemini 2.5 Flash Image says the model works best with no more than 3 input images, so a smaller, purposeful reference set is the safer starting point.

Legacy status does not mean the Kyeo route is offline

Treat lifecycle status as one selection factor, not a prediction of any single output, and validate important work with your own material.

Capabilities

Text-to-image generation

When no image is attached, Kyeo sends the prompt to `google/nano-banana`. You can choose an output format and aspect ratio, and the request costs a fixed 3 credits.

Natural-language image editing

Attaching an image switches the request to `google/nano-banana-edit`. You can describe what to preserve, remove, replace, or restyle, but each output still needs a visual check for instruction adherence.

Format, ratio, and input boundaries

The current workflow offers PNG or JPEG output and 11 aspect ratios. Technical identifiers: `image_urls`.

Use cases

Create a new image from a brief

Describe the subject, setting, composition, and visual treatment, then choose a landscape, portrait, or square aspect ratio for the first draft.

Make a targeted change to an existing image

Try adding or removing an object, replacing a background, adjusting color, or changing style. Manually verify untouched details when identity, product accuracy, or brand rules matter.

Compose from a small reference set

Provide separate references for the subject, material, or scene and name each role in the prompt. Begin with 1 to 3 images instead of treating the interface maximum as a target.

Prompt tips

Lead with the subject and the requested action

State what to create or change in one sentence, then list what must remain, what to replace, and what to avoid. Actions are easier to verify than a loose string of style adjectives.

Label the purpose of every reference

Identify which image supplies the subject, material, pose, or background, and start with 1 to 3 inputs. As the set grows, check for mixed-up or omitted elements.

Choose the ratio and format before submitting

Match 1:1, 16:9, 9:16, or another ratio to the intended placement, then select PNG or JPEG. Review text, hands, edges, subject features, and areas that were not meant to change.

Strengths

Text-to-image generation and editing are available from the same page.

Both routes cost 3 credits per request, making the cost predictable.

Prompts allow 5,000 characters and the form offers 11 aspect ratios.

The editing route can accept multiple images with separately defined roles.

Limitations

Google now lists Gemini 2.5 Flash Image as a legacy model, so long-lived workflows should also assess newer options.

The current form has no 1K, 2K, or 4K output tier and no search grounding.

The 10-image upload limit is capacity, not a recommendation or a consistency guarantee.

Text, identity, product details, consistency, and rights still require human review.

Adjustable parameters

Quickly assess whether this model matches your workflow.

Output format

output_format

Optional

Parameter type: select

Default: png

PNG

JPEG

Aspect ratio

image_size

Optional

Parameter type: select

Default: 1:1

1:1

9:16

16:9

3:4

4:3

3:2

11 choices

Credit usage

3 credits per request

A text-to-image or image-editing request currently costs a fixed 3 credits on Kyeo AI.

Budget tip

Before batch generation, run an A/B test with the same assets across the current model and alternatives to validate quality and avoid wasting credits.

FAQ

Related models

Compare these similar candidates before deciding.

Nano Banana Pro

Nano Banana Pro supports text-to-image generation and reference-image editing on Kyeo AI, with 1K, 2K, or 4K output. Each 1K or 2K request costs 8 credits, while 4K costs 14. Resolution is an output-size tier, not a guarantee of quality, accuracy, or detail.

AI image generation and editing model

8–14 credits per request

View details Use now

Nano Banana 2

Kyeo AI offers generation and reference-guided editing through `nano-banana-2`, with optional `google_search`, up to 14 `image_input` files, 15 aspect-ratio choices, and 1K, 2K, or 4K output. Those tiers currently cost 5, 8, and 12 credits, and prompts allow 20,000 characters.

AI image generator and editor

5–12 credits per request

View details Use now

GPT Image 1.5

GPT Image 1.5 supports generation and editing on Kyeo AI. Text-only requests use `gpt-image/1.5-text-to-image`; attaching references switches to `gpt-image/1.5-image-to-image`. Kyeo exposes 3 aspect ratios and `medium / high` tiers at 4 / 22 credits. OpenAI now marks GPT Image 1.5 as a deprecated previous model, so new work should also benchmark a current model such as GPT Image 2.

AI image model

4-22 credits / request

View details Use now

Sources

Content is based on model documentation, feature references, and the settings available on Kyeo AI.

Source notes

Kyeo calls `google/nano-banana` when no image is attached and `google/nano-banana-edit` when images are supplied. 10 is an interface limit, not a recommendation or consistency guarantee.

Last reviewed: 2026-07-20

Google AI for Developers: Nano Banana

Official

View source

Google AI for Developers: Image generation with Gemini

Official

View source

Upstream interface documentation: Google Nano Banana

Platform

The upstream text-to-image documentation lists `google/nano-banana`, a 5,000-character prompt limit, PNG or JPEG output, and 11 aspect ratios.

Upstream interface documentation: Nano Banana image editing

Platform

This is an interface limit, not an output guarantee. Technical identifiers: `google/nano-banana-edit`, `image_urls`.

Current Kyeo credit configuration

Platform

Kyeo currently charges 3 credits for either a Nano Banana generation or editing request.

Kyeo AI model configuration

Platform

Kyeo switches between the generation and editing model IDs based on image input, with the current form and runtime configuration defining the available controls.