Quick wins for a faster PC:
Repair Windows errors before they cause bigger problemsFix Now →Scan for outdated or missing drivers - takes under a minuteDriver Scan →Clear out junk files and repair common Windows errorsFree Scan →
ThatPainter is reader-supported. When you buy through links on our site, we may earn an affiliate commission. Learn More
Nano Banana is Google’s name for Gemini’s image-generation and editing capabilities, not a separate AI website. The original viral model was Gemini 2.5 Flash Image; Google’s current documentation lists it as a legacy model alongside newer options, including Nano Banana 2.1, Nano Banana 2 Lite, Nano Banana 2, and Nano Banana Pro. The viral examples show image edits, multi-image compositions, style changes, and visual storytelling—but they are Google demonstrations, not independent performance tests.
For painters and visual artists, the tools can help explore composition, style, and visual ideas. Treat their outputs as generated imagery: likeness, text, and factual details may change or be wrong.
What is Nano Banana?
“Nano Banana” began as a public codename for a Gemini image model. Google now uses the name for Gemini-native image generation and editing, including conversational work from text, images, video, or combinations of inputs. Because the name covers a changing family of models, it is useful to distinguish the original viral model from current options.
| Public name | Documented model ID | What Google describes it for |
|---|---|---|
| Nano Banana 2.1 | gemini-nano-banana-2.1 | Updated high-efficiency generation and conversational editing, with improved quality, text rendering, multi-turn consistency, and Search grounding. |
| Nano Banana 2 Lite | gemini-3.1-flash-lite-image | Fast, low-cost generation; not optimized for multiple references or multiple sequential edits. |
| Nano Banana 2 | gemini-3.1-flash-image | Previous-generation workhorse; Google recommends Nano Banana 2.1 for new projects. |
| Nano Banana Pro | gemini-3-pro-image | Complex visual tasks, world knowledge, localization, brand consistency, and precision control. |
| Original Nano Banana | gemini-2.5-flash-image | The legacy model associated with the original viral release. |
These names and descriptions reflect Google’s current developer documentation as researched on October 7, 2026. Features and model choices may differ between Gemini Apps and the developer API, and can change over time.
#1 Best Overall
Why did Nano Banana go viral?
Its viral appeal came from more than turning text into pictures. In Google’s September 12, 2025 examples article, contributing writer TK Breuer described the model’s ability to keep a person or pet consistent across edits, blend photos, change parts of an image while retaining other areas, and apply a style while preserving important details. That is Google’s description of its own tool, not an independent test.
The examples were easy to share because they showed recognizable transformations in a single image. They also demonstrated a conversational workflow: provide an image or images, request a change, and refine the result.
10 Nano Banana examples and what they show
Google’s September 2025 article promises ten examples, but the accessible text in the research for this article exposed nine prompt blocks. The nine examples below are those recovered from that article. The tenth is identified separately as an example from Google’s editing playbook rather than being presented as part of the nine recovered prompts.
Free tools Windows power users keep installed
One-click scans. No signup required.
1. Turn a cat into a 16-bit game character
Google’s example asks for a cat to become a 16-bit video-game character in a platform-game level. It illustrates a style transformation that also places the subject in a new visual setting.
2. Bring a person and her childhood self together
One prompt combines an adult subject with a childhood version of herself at a tea party. It demonstrates multi-image composition and the challenge of retaining recognizable details across references.
3. Turn a landscape into a fantasy map
A landscape photo becomes a stylized 3D world map for a fantasy game. It shows how an existing image can serve as the basis for a new visual interpretation.
4. Convert a photo into a pencil drawing
This style-transfer prompt asks for a photo to be rendered as a pencil drawing. For painters, a transformation like this can suggest how a scene might read in a different medium, but it is not a substitute for observation or a guarantee of a particular artistic result.
5. Reimagine ingredients as a dessert
The example turns ingredients into a refined dessert plated in an avant-garde restaurant style. It combines subject transformation with a specified presentation.
6. Turn scissors into a fantasy character
This imaginative prompt transforms an ordinary object into a character. It illustrates concept generation rather than a realistic product depiction.
7. Depict a dog as a packaged figurine
A dog is shown as a small 3D figurine with gift packaging. The packaging and toy are generated parts of the image, not products required to use Gemini or recommendations for real merchandise.
8. Create a nine-image superhero story
The prompt requests a nine-image superhero story with two protagonists. It demonstrates multi-image storytelling and consistency across a sequence; inspect any generated lettering and continuity before sharing.
PC Slower Than It Used to Be?
A free scan shows the junk files, broken settings and background clutter dragging Windows down - then fixes them in one click.Free scan · Windows 10 & 11Crashes, No Sound, or Screen Glitches?
Random freezes, missing sound and display glitches usually trace back to one bad driver. Find and replace yours safely.Free scan · under a minute9. Change a dress into tennis balls
This unusual transformation asks for a person’s dress to be made of tennis balls. It is a playful example of changing part of an image while retaining the broader scene.
10. Try a hairstyle edit
Google’s separate Gemini image-editing playbook includes a prompt to create a new hairstyle using an uploaded selfie. Ask for the hair change while specifying which features—such as the face, glasses, expression, and framing—should remain the same. Those constraints are requests, not guarantees.
How to use Nano Banana in Gemini
Google’s Gemini Apps help describes opening the Images area, choosing a template or writing a prompt, then editing a generated image or uploading one. Users can also upload several images and ask Gemini to create a new image based on them.
- Open the Images area in Gemini.
- Choose a template or write a prompt describing the image or edit.
- Upload one or more reference images if needed.
- State what should change and what details should be preserved.
- Review the result and request a targeted revision if needed.
- Check faces, hands, text, factual claims, and unintended changes before sharing or using the image.
Google says availability, quotas, country support, and account access can vary. Editing is not available to users under 18; some connected-photo features have additional age and geography restrictions. Pro regeneration is available to eligible paid subscribers, according to Google’s help page. Check current Gemini help for your region and account rather than relying on a universal price, quota, or access claim.
Do these 3 things before closing this tab:
1Clear out junk files and repair common Windows errors2Scan for outdated or missing drivers - takes under a minute3Repair Windows errors before they cause bigger problemsUsing Nano Banana through AI Studio or the Gemini API
Google’s developer documentation points developers to AI Studio and the Gemini API. It describes Nano Banana 2.1 as the recommended option for new projects, while Nano Banana 2 Lite is designed for speed and lower cost, Nano Banana 2 is its previous-generation workhorse, Nano Banana Pro is aimed at demanding visual tasks, and the original Gemini 2.5 Flash Image is legacy.
Choose based on the task and the current API documentation. The model names, IDs, and feature descriptions can change; do not assume that consumer Gemini access and API access offer the same model selection or limits. Google says generated images include a SynthID watermark.
Prompting patterns for more useful results
A clear prompt can separate the requested change from the details you want retained:
- Subject: Identify the person, pet, object, or scene.
- Change: State the transformation, such as a new style, setting, or hairstyle.
- Constraints: Specify what should stay recognizable or unchanged.
- Style and layout: Describe the visual direction or sequence, if relevant.
- Checks: Ask for legible text or consistent details when they matter, then review the result yourself.
For revisions, request one change at a time. Even when you ask the model to keep everything else the same, inspect the whole image: a generative edit can change areas beyond the requested one.
The Tool Desk
Outbyte Driver Updater FREEFix the driver behind crashes, sound loss and screen glitchesFind Drivers →Outbyte PC Repair FREERepair Windows errors before they cause bigger problemsFix Now →What Nano Banana does not prove
A realistic-looking image is not proof of an event
A convincing photograph-like image does not establish that the depicted scene happened. Do not treat generated imagery as proof of identity, historical evidence, or documentation without independent verification.
Identity preservation is not guaranteed
The model may alter a face or other details despite instructions to preserve them. Obtain appropriate permission before using another person’s likeness, and consider privacy when uploading personal or sensitive images.
Polished text and visuals can still be wrong
Generated labels, captions, and other text can contain errors. An attractive educational or documentary image can also include inaccurate details. Verify important claims independently.
Generated edits still involve rights and responsibilities
Google says users must follow its Terms of Service and Prohibited Use Policy, including respecting copyright and privacy rights. Before uploading a family photo, a client portrait, a child’s image, or somebody else’s artwork, consider whether you have permission and whether the edit could mislead viewers.
How SynthID helps—and what it cannot tell you
Google says its generated images include an invisible SynthID mark; its consumer overview also describes a visible AI-generated marker. Gemini’s verification feature can check whether supported media was generated or edited by Google AI. These are Google’s stated provenance features, not a guarantee that every copied, transformed, or exported image will retain every indicator in every context.
A positive check can be a provenance clue. A negative result does not prove that an image was made by a human or that another AI system was not used.
A sensible artist’s workflow
- Use your own references or get permission. Keep track of where source images came from.
- Start with exploratory prompts. Use the tool to test compositions, styles, and ideas rather than treating the first output as finished work.
- Keep references distinct from finished work. Label generated studies clearly in your project files.
- Inspect the result closely. Look for changed facial features, distorted anatomy, broken lettering, and other unintended edits.
- Fact-check educational or documentary content. A polished image is not verification.
- Disclose important transformations. This is especially relevant for restoration, portraits, historical-looking scenes, and editorial work.
- Limit sensitive uploads. Do not upload private material merely to test a novelty effect.
For painters, a useful role for the model is often to explore possibilities—such as alternative arrangements or visual styles—while leaving artistic judgment and the final decisions to the artist.
FAQ
What is Nano Banana AI?
Nano Banana is Google’s name for Gemini-native image generation and editing. The original viral model was Gemini 2.5 Flash Image; Google’s current documentation lists newer models in the same family.
Recommended Free Tools
How do I use Nano Banana in Gemini?
Open the Images area in Gemini, choose a template or enter a prompt, and optionally upload one or more images to edit or combine. Access and features can vary by account and region.
Can Nano Banana edit a photo and keep the person looking the same?
You can ask it to preserve a person’s recognizable features while changing other parts of a photo, but likeness preservation is not guaranteed. Check the result and get permission before using someone else’s image.
What are the viral Nano Banana photo trends?
Google’s examples include turning a cat into a game character, combining an adult with her childhood self, making a fantasy map from a landscape, creating a figurine-style dog image, and generating a multi-image superhero story.
Is Nano Banana the same as Gemini 2.5 Flash Image?
Gemini 2.5 Flash Image was the original model associated with the viral Nano Banana name. Google now uses Nano Banana for a changing family of image-generation and editing models, so the name does not refer only to that original model.
Quick wins for a faster PC:
Repair Windows errors before they cause bigger problemsFix Now →Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Clear out junk files and repair common Windows errorsFree Scan →Can SynthID prove that an image is real?
No. SynthID can provide a provenance clue for supported media generated or edited by Google AI. A negative result does not prove that an image is human-made or that no other AI system was used.
The Bottom Line
Nano Banana is Google’s branding for Gemini image generation and editing, and the original Gemini 2.5 Flash Image model is now part of a wider, changing family. Google’s viral examples show creative transformations, photo edits, and multi-image storytelling; they are demonstrations, not independent performance tests. Use the tool for exploration, check its outputs carefully, and consult current Google documentation for model names and access details.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
