How to Use Nano Banana: Step-by-Step in the Gemini App, Google Slides & Vids
Nano Banana is the nickname for Google's Gemini image models — the ones that generate a picture from a sentence, or take a photo you upload and change just the part you asked about. This guide shows you exactly how to use Nano Banana in the three places most people meet it: the Gemini app, Google Slides, and Google Vids. Every method is a numbered list you can follow along with in another tab.
If you want the background first, our What is Nano Banana? explainer covers where the name came from and how the model versions differ. This article is the hands-on part: open the tool, type the prompt, get the image.
A quick note on accuracy: Google updates Gemini frequently, and menu labels, model names, and plan availability can shift between the time we write this and the time you read it. The steps below describe the flow generically where labels change often, and we tell you when to double-check something on Google's own site.
What you need before you start
- A Google account. That is all you need for the Gemini app. A free account gets you image generation with daily limits; paid Google AI plans raise those limits and unlock the Pro model.
- A Google Workspace Business or Enterprise plan if you want to use Nano Banana inside Google Slides and Google Vids. Gemini has been included in those plans since January 2025 with no separate AI add-on. See our Google Workspace Gemini guide for what each tier includes, or the pricing page for current costs.
- A clear idea of what you want. Not a design brief — just a sentence. “A product photo of a ceramic mug on a wooden desk, morning light” is plenty.
- Optionally, a reference photo. If you are editing an existing image or want a specific person, product, or logo to stay consistent, have that file ready to upload.
Which Nano Banana model am I using?
There are three generations, and the name you see in the app depends on your plan and the date:
- Nano Banana (Gemini 2.5 Flash Image, August 2025) — the original fast model. Great for everyday edits and quick concepts.
- Nano Banana Pro (Gemini 3 Pro Image, November 2025) — higher fidelity, up to 4K output, and far better at rendering readable text inside the image. This is the one powering Slides.
- Nano Banana 2 (Gemini 3.1 Flash Image, February 2026) — the newer fast model with much of the Pro quality at Flash speed.
For most of this tutorial the steps are identical regardless of version. Where the model choice matters (readable text, fine detail), we point it out.
Method 1: How to use Nano Banana in the Gemini app
This is the simplest route and the one that works with a free Google account. It is the same on the web at gemini.google.com and in the Gemini mobile app.
- Open Gemini. Go to gemini.google.com in a browser, or open the Gemini app on your phone, and sign in with your Google account.
- Start a new chat. A fresh conversation keeps the model focused on your image and makes later edits cleaner.
- Choose the image model (if your plan shows a picker). Depending on your plan, there is a model or tools menu near the prompt box where you can select Nano Banana or Nano Banana Pro. If you do not see a picker, just describe the image — Gemini routes image requests to the image model automatically. Exact menu labels vary, so check Google's help pages if you are unsure.
- Type your prompt. Describe the subject, the setting, the style, and the lighting in plain language. Example: “A flat-lay photo of a leather notebook, fountain pen, and espresso cup on a white marble table, soft window light, square format.”
- Or upload a photo and describe the edit. Use the attach (plus or paperclip) button to add an image, then tell Gemini what to change: “remove the background,” “put this product on a marble table,” “make it a watercolor.” You can upload several images and ask Gemini to combine them — for example, a product photo plus a background photo.
- Send and wait a few seconds. The image appears in the chat. Fast models return in a handful of seconds; Pro takes a bit longer.
- Refine with follow-up messages. This is the part that makes Nano Banana feel different. Reply in the same chat: “make the lighting warmer,” “change the mug to navy blue,” “same scene, but from a lower angle.” Multi-turn editing keeps the subject consistent instead of generating something new from scratch.
- Download. Hover over (or tap) the image and use the download option to save it. On mobile you can also share directly to other apps.
Three example prompts to try first
- “A minimalist logo-style illustration of a fox reading a book, flat vector style, two colors: forest green and cream, white background.”
- “A realistic photo of a small bakery storefront at golden hour, striped awning, chalkboard sign that says ‘Fresh Sourdough Daily’, 16:9.” (Text in quotes renders far better on Nano Banana Pro.)
- Upload a headshot, then: “Keep my face and pose identical. Replace the background with a softly blurred modern office and change my shirt to a plain navy crew neck.”
Want more starting points? Our Nano Banana prompt library has dozens sorted by use case.
Method 2: How to use Nano Banana in Google Slides
Inside Slides, Nano Banana works through the Gemini features built into Workspace. Since November 2025 the image generator in Slides has been powered by Nano Banana Pro, which is why it is now genuinely useful for infographics and slides where the text has to be legible. You need a Workspace Business or Enterprise plan (or a compatible Google AI plan) with Gemini enabled by your admin.
- Open a presentation in Google Slides and go to the slide where you want the image.
- Open the image generator. There are two ways in: click the Gemini icon (top right) to open the side panel and ask it to create an image, or go to Insert → Image and choose the option to create or visualize an image with AI. The exact wording has changed a few times (“Help me visualize” and “Help me create an image” are both used); look for the sparkle icon.
- Describe the image. Same rules as the app: subject, setting, style. For slides, say how it will be used — “a wide 16:9 hero image for a title slide about renewable energy, clean and modern, lots of negative space on the left for text.”
- Pick a style if the panel offers one (photography, illustration, sketch, watercolor, and so on), then generate.
- Choose from the results and insert. Slides usually shows several options. Click the one you like to place it on the slide, then resize and position it like any other image.
- Regenerate if needed. Adjust your description and run it again. It is normal to take two or three passes.
- Clean up inserted images. Select any image on a slide and use the remove-background option in the image toolbar or side panel to cut it out for a cleaner layout.
For a deeper walkthrough of the Slides side specifically, including template and branding tips, read Gemini in Google Slides: Help Me Visualize.
Method 3: How to use Nano Banana in Google Vids
Google Vids is Workspace's video builder, and Nano Banana is how you create the still images that become scene backgrounds, product shots, and illustrations inside it. The flow is conversational, like the Gemini app, but it lives inside the Vids editor.
- Open or create a video at vids.google.com (or from Drive → New). You can start from a blank project, a template, or let Gemini draft a storyboard from a prompt.
- Select the scene that needs an image.
- Open the image generation tool from the media or Gemini panel and choose the option to generate an image.
- Describe the visual for that scene. Think in terms of video: “a wide, calm shot of a sunrise over a city skyline, cinematic, muted blues and oranges, room for a title on the left.”
- Refine with follow-ups. Vids supports multi-turn editing, so you can say “make it dusk instead” or “add a subtle lens flare” and keep the composition.
- Place the image in the scene, then layer text, voice over, or your own footage on top as usual.
- Repeat per scene, reusing the same descriptive language (colors, mood, style) so the video feels consistent from start to finish.
We cover the rest of the Vids workflow — scripts, AI voice over, and stock media — in Google Vids: AI video for teams.
How to edit an existing photo with Nano Banana
Editing is where Nano Banana really earns its reputation. Rather than masking and cloning by hand, you upload a picture and say what should change — and, just as importantly, what should stay the same. These steps are for the Gemini app, which gives you the most control.
- Start a new chat and upload the photo. Use the attach button. Higher-resolution originals give better results.
- State what to keep. Lead with the anchor: “Keep the person's face, hair, and pose exactly the same.” This single line prevents most unwanted changes.
- Describe one change. “Change the outfit to a navy suit with a white shirt.” One edit per message is more reliable than five at once.
- Review the result closely. Zoom in on hands, edges, and any text. If something drifted, say so directly: “The face changed — restore the original face.”
- Stack further edits in the same chat. “Now remove the background and replace it with plain white.” Then: “Add a soft shadow under the subject.” Each turn builds on the last.
- Combine images if needed. Upload a second photo (a product, a logo, a location) and ask Gemini to place it into the first: “Put the product from image 2 on the table in image 1, matching the lighting.”
- Download the final version and keep your original — it is worth re-running an edit from the source if you change your mind later.
There is a full deep-dive with before-and-after examples in Nano Banana image editing: the complete guide.
10 quick Nano Banana prompt recipes
Copy, paste, and swap in your own details. Anything in single quotes is text you want rendered inside the image — use Nano Banana Pro for those.
- Product on a surface: “Studio photo of [product] on a white marble surface, soft diffused light, slight reflection, square format for Instagram.”
- Background swap: “Keep the subject identical. Replace the background with a bright, minimal Scandinavian living room, shallow depth of field.”
- Headshot polish: “Keep face and expression exactly the same. Even out the lighting, neutral grey backdrop, professional LinkedIn style.”
- Style transfer: “Turn this photo into a loose watercolor painting, visible brush strokes, paper texture, keep the composition.”
- Slide hero image: “16:9 abstract background for a slide about cybersecurity, dark navy with thin glowing lines, empty space on the right for text.”
- Infographic (Pro): “A clean three-step infographic titled ‘How We Onboard’ with steps ‘Kickoff’, ‘Setup’, ‘Launch’, flat icons, brand colors teal and charcoal, 16:9.”
- Social post with text (Pro): “Square announcement graphic, bold headline ‘Summer Sale — 20% Off’, sunny beach photo background, modern sans-serif type, high contrast.”
- Object removal: “Remove the people in the background and fill the area naturally. Keep everything else unchanged.”
- Colorway variations: “Show this same sneaker in three colorways side by side: all black, white with red accents, and olive green.”
- Sketch to render: “Turn this rough pencil sketch of a chair into a realistic product render, light oak and black steel, white background.”
Tips for better Nano Banana images
- Use the five-part formula. Subject + setting + style + lighting + composition. You do not need all five every time, but if a result feels off, the missing piece is usually one of these.
- Put rendered text in quotes. “A sign that says ‘Open Late’” is far more reliable than “a sign about being open late.” Keep it short and use the Pro model.
- Say what to keep before what to change. For edits, the “keep” instruction is the most important sentence in the prompt.
- Name the aspect ratio or destination. “Square for Instagram,” “16:9 for a slide,” “vertical 9:16 for a story.” Gemini will usually match it.
- Iterate in the same chat. Small follow-up nudges beat rewriting the whole prompt. If the conversation gets long and results start drifting, start a fresh chat and re-upload your best version.
- Use a reference image for anything specific. A real product, a real logo, a particular room — describing them never works as well as showing them.
- Ask for options. “Give me three variations with different color palettes” is a fast way to explore.
- Be plain, not poetic. Nano Banana reads natural language well. Clear beats clever.
Common problems and how to fix them
The text in my image is garbled
This is the most common complaint with the original Flash model. Switch to Nano Banana Pro (Gemini 3 Pro Image), keep the text short, put it in quotes, and specify the font style (“bold sans-serif, all caps”). In Slides, you are already on Pro, so shorten the text and try again.
It generated the wrong subject or changed my subject
Upload a reference image rather than describing the subject, and open the prompt with an explicit keep instruction (“keep the face and pose identical”). If a specific detail keeps drifting, name it directly and ask for just that fix.
Hands, edges, or reflections look off
Ask for a targeted correction in the next message: “fix the left hand so it has five fingers,” “clean up the edge around the hair.” If it persists, regenerate with a simpler composition or crop tighter.
I hit a usage limit
Free Gemini accounts have a daily cap on image generations; paid Google AI plans and Workspace tiers raise it. Limits change, so check the current numbers on Google's site. Batch your edits into fewer, clearer prompts to stretch what you have.
I don't see the image option in Slides or Vids
Either your plan does not include Gemini, or your Workspace admin has not turned it on. Business Starter and above include Gemini; check with your admin, or compare tiers on our Gemini in Workspace page.
A note on responsible use
Every Nano Banana image carries an invisible SynthID watermark that identifies it as AI-generated. Do not generate images of real people without their consent, and review anything you plan to use commercially — check the details, check the licensing terms of any reference material, and check Google's current usage policies.
Frequently asked questions
Is Nano Banana free to use?
Yes, with limits. Anyone with a Google account can generate and edit images with Nano Banana in the Gemini app for free, subject to a daily cap. Paid Google AI plans raise the limits and unlock Nano Banana Pro. Using it inside Google Slides and Google Vids requires a Google Workspace Business or Enterprise plan, where Gemini is included at no extra cost. Check Google’s site for current limits.
How do I edit a photo with Nano Banana?
Open the Gemini app, start a new chat, upload the photo with the attach button, and describe the change in plain language. Start by saying what should stay the same ("keep the face and pose identical"), then describe one edit at a time. Continue in the same chat to stack further edits, then download the result.
What is the difference between Nano Banana and Nano Banana Pro?
Nano Banana (Gemini 2.5 Flash Image) is the fast, everyday model. Nano Banana Pro (Gemini 3 Pro Image) produces higher-fidelity images up to 4K and is much better at rendering readable text, which makes it the right choice for infographics, posters, and slides. Nano Banana 2 (Gemini 3.1 Flash Image) is the newer fast model that closes much of the quality gap.
Can I use Nano Banana in Google Slides?
Yes. In Google Slides, open the Gemini side panel or go to Insert, then Image, and choose the AI image option (labelled "Help me visualize" or "Help me create an image" depending on your version). Describe the image, pick a style, and insert it. This requires a Workspace plan with Gemini enabled, and it has been powered by Nano Banana Pro since November 2025.
Why is the text in my Nano Banana image misspelled?
Text rendering is the weak spot of the original Flash model. Switch to Nano Banana Pro, keep the text short, put the exact words in quotes in your prompt, and specify a font style such as "bold sans-serif". If it still drifts, regenerate or ask for just the text to be corrected in a follow-up message.
Can I use Nano Banana images commercially?
Generally yes, subject to Google’s terms and usage policies, which you should review before commercial use. Every image carries a SynthID watermark identifying it as AI-generated. Avoid generating images of real people without consent, and be careful with reference images you do not own the rights to.
Does Nano Banana work on my phone?
Yes. The Gemini mobile app on Android and iOS supports the same image generation and editing flow as the web: start a chat, type a prompt or attach a photo, describe the edit, refine with follow-ups, and save or share the result.
Next steps
The fastest way to learn how to use Nano Banana is to open the Gemini app right now, upload a photo, and type one sentence. Once the basic loop clicks — describe, generate, refine, download — the Slides and Vids versions feel like the same tool in a different room.
If your team is on free Gmail and wants Nano Banana inside Slides and Vids, a Google Workspace Business plan is the route, and Gemini is included at no extra cost. See current pricing or jump straight to the trial.
Related reading: What is Nano Banana? · Nano Banana Prompts · Nano Banana Image Editing · Gemini in Google Slides · Google Vids AI Video