Nano Banana 2.1 Prompts: Infographics, Panoramas, Text Edits
10+ copy-paste Nano Banana 2.1 prompts for infographics, 4:1 panoramas, sign text, photo merges and consistent characters, with credits.
Updated Nov 19, 2026 8 min read

Nano Banana 2.1 (Google's newest image model, in Plurel as nano-banana-2) is built for exactly the things most AI image models get wrong: small readable text, dense infographics, wide panoramas, and a character that stays the same from one picture to the next. The prompts below are copy-paste ready, with the aspect ratio, resolution and credit cost to pick for each.
What Nano Banana 2.1 is actually good at
In Plurel's model list, Nano Banana 2.1 is described as Google's newest photorealistic model, built for sharper text, infographics and steady characters. That's the model to reach for when a prompt needs exact words on a sign, a multi-line list of labels, or a face that has to look the same three images later — not for loose, painterly or highly stylised work, where Seedream 5 or Qwen Image 3 often do better.
Every image from this model costs 4 credits, whether you're generating from scratch or editing an existing picture.
How to write prompts that keep the text right
Three habits matter more than clever wording:
- Put the exact words in quotation marks. Write
reading "TODAY'S SPECIAL"rather than just "a special sign" — the model copies quoted text far more reliably than paraphrased text. - Describe layout, not just content. Say where each line sits (top, left column, under the headline) instead of listing text and hoping the model arranges it well.
- One visual idea per generation. A prompt asking for a five-point infographic and a cartoon mascot and a background photo tends to lose accuracy on all three. Build the base image first, then add details with a follow-up.
Do it in Plurel
- Open the chat and click the Image chip under the composer, or go straight to plurel.tech/?model=nano-banana-2.
- Pick Nano Banana 2.1 from the model picker if it isn't already selected.
- Set the aspect ratio (square, portrait, landscape, or the wide 4:1 option for panoramas) and the resolution (1K for quick drafts, 2K for most posts and signs, 4K for posters or anything you'll print or crop).
- Paste the prompt, generate for 4 credits, and check the text against your prompt before you move on.
- To change anything — colour, wording, pose, time of day — just reply in the same chat. It edits the image that's already on screen, so you don't need to re-describe the whole scene each time.
- To merge a product or person into an existing photo, open Edit instead, attach both pictures, and describe how they should combine. Same 4-credit price.
Infographic prompts
A clean vertical infographic titled "5 Steps to Better Sleep." Five numbered sections stacked top to bottom, each with a small icon on the left and a short headline on the right: "1. Fixed wake time", "2. No caffeine after 2pm", "3. Dark, cool room", "4. No screens in bed", "5. Wind-down routine". White background, navy headlines, one teal accent color for the icons and numbers. Sans-serif font throughout.
A numbered 5-step list graphic with the exact wording in place. Aspect ratio: 9:16. Resolution: 2K. Cost: 4 credits.
A side-by-side comparison graphic with two columns labeled "Plan A" and "Plan B". Under each column, three rows of text: "Storage: 100GB" vs "Storage: 1TB", "Price: $5/mo" vs "Price: $9/mo", "Support: Email" vs "Support: 24/7 Chat". Simple table lines, white background, bold column headers.
A two-column pricing comparison with every label spelled out. Aspect ratio: 16:9. Resolution: 2K. Cost: 4 credits.
A square Instagram carousel slide titled "3 Mistakes New Runners Make" with three bullet points below it: "• Increasing mileage too fast", "• Skipping warm-ups", "• Wearing worn-out shoes". Minimal flat design, one bold accent color, plenty of white space.
A social-ready bullet-list slide. Aspect ratio: 1:1. Resolution: 2K. Cost: 4 credits.
A simplified top-down map of a small town with five labeled pins: "Town Hall", "Library", "Main Street Market", "River Park", "Train Station", connected by a dotted walking path. Flat illustrated map style, muted green and beige palette.
A labeled map diagram, useful for event or tourism graphics. Aspect ratio: 1:1. Resolution: 2K. Cost: 4 credits.
Panorama prompts
A 4:1 ultra-wide panorama of a mountain lake at golden hour, pine forest on both shores, one small wooden dock extending into still water, mist rising off the surface, warm orange light on the peaks, cool blue shadows in the water, photorealistic.
A banner-width landscape shot that works as a header image. Aspect ratio: 4:1. Resolution: 4K (the extra width needs the detail). Cost: 4 credits.
A 4:1 ultra-wide panorama of a city skyline at blue hour, glass towers reflecting pink and orange sky, a river in the foreground with one bridge crossing it, lights just starting to switch on in the windows.
A wide skyline banner for a website hero or desktop background. Aspect ratio: 4:1. Resolution: 4K. Cost: 4 credits.
Chalkboard and sign text prompts
A rustic café chalkboard sign leaning against a brick wall, hand-chalked white lettering reading "TODAY'S SPECIAL: Pumpkin Spice Latte — $4.50", a small chalk drawing of a coffee cup in the corner, warm string lights blurred in the background, shallow depth of field.
A photo-real chalkboard with the exact price and wording. Aspect ratio: 4:3. Resolution: 2K. Cost: 4 credits.
A vintage neon shop sign mounted above a narrow doorway reading "BLUE MOON BOOKS" in cursive neon tubing, warm pink and blue glow, rain-slicked street reflecting the sign at night, cinematic photo.
A neon storefront sign for a brand mockup or mood board. Aspect ratio: 3:4. Resolution: 2K. Cost: 4 credits.
A movie-style poster for a fictional film called "THE LAST LIGHTHOUSE", with the title in tall bold serif letters at the top, the tagline "Some storms never end" in smaller italics beneath it, and the text "IN THEATERS JUNE 12" at the bottom. A lone lighthouse on a cliff against a stormy sky fills the background.
A three-block text poster — title, tagline, date — all legible. Aspect ratio: 2:3. Resolution: 4K. Cost: 4 credits.
Combining two photos
Open Edit, attach your two source photos, and describe the merge in one sentence:
Place the sneaker from the first photo onto the marble countertop in the second photo, matching the lighting and shadow direction of the room, keep the sneaker's exact colors and logo.
Drops a product photo into a real room shot without a separate photoshoot. Aspect ratio: match the scene photo. Resolution: 2K. Cost: 4 credits.
Combine the dog from the first photo with the beach background from the second photo, standing on the sand facing the water, matching the warm late-afternoon light of the beach photo.
Merges a pet photo with a new background while keeping the pet recognisable. Aspect ratio: match the background photo. Resolution: 2K. Cost: 4 credits.
Recolouring text in an edit
Don't regenerate the whole image for a small text change — edit it instead:
Change the text on the sign from red to deep teal. Keep the font, size, and position exactly the same, and leave everything else in the photo untouched.
Swaps a text color while leaving the rest of the picture alone. Aspect ratio: same as the original. Resolution: same as the original. Cost: 4 credits.
Keeping one character across follow-ups
Generate the character once, then keep editing the same chat instead of writing a fresh prompt each time — Plurel's chat edits whatever was just made, so the face, colors and outfit carry over automatically.
A friendly red fox wearing round glasses and a tiny green scarf, standing upright like a shopkeeper, photorealistic fur detail, soft studio lighting, plain cream background.
The base character. Aspect ratio: 4:3. Resolution: 2K. Cost: 4 credits.
Same fox, now sitting behind a wooden counter stacked with jars of honey, same glasses and scarf, same lighting.
First follow-up — new scene, same character. Aspect ratio: 4:3. Resolution: 2K. Cost: 4 credits.
Same fox, now outside at night in front of the shop under string lights, same glasses and scarf.
Second follow-up — new time of day, same character. Aspect ratio: 4:3. Resolution: 2K. Cost: 4 credits.
Quick reference
| Prompt type | Aspect ratio | Resolution | Cost |
|---|---|---|---|
| Infographic / list | 1:1 or 9:16 | 2K | 4 credits |
| Comparison table | 16:9 | 2K | 4 credits |
| Panorama / banner | 4:1 | 4K | 4 credits |
| Chalkboard / sign | 3:4 or 4:3 | 2K | 4 credits |
| Poster | 2:3 | 4K | 4 credits |
| Photo merge (Edit) | match source | 2K | 4 credits |
| Character follow-up | match first image | 2K | 4 credits |
Troubleshooting
If a word comes out misspelled, don't retype the whole prompt — reply with just the fix, for example "fix the spelling of 'Lighthouse' in the title," and let the follow-up correct it in place. If a panorama looks squeezed or oddly stretched, drop the resolution from 4K to 2K and regenerate; very wide ratios at the top resolution occasionally overdo the stretch. If a character drifts after several follow-ups (different fur color, missing scarf), start a new chat from the original image and describe the fixed details again — long edit chains lose a little consistency, and a fresh start is faster than arguing it back into shape. For text accuracy generally, short phrases in quotes beat long sentences, and all-caps signage tends to render more reliably than mixed case.
For more on why AI models fumble text and faces in the first place, see Why AI Video Looks Wrong (the same rendering issues show up in stills) and How AI Image Generators Work. For keeping a face steady across a whole project rather than one follow-up chain, see Consistent Character Prompt and How to Keep the Same Character Across AI Images. If you're deciding between Nano Banana 2.1 and other text-heavy models, Nano Banana vs GPT Image 2.6 and Nano Banana 2.1: What Google Improved and How to Use It cover the differences.
Start with one prompt from each section above, check the text against what you typed, and build from there — that's the whole workflow.
Questions people ask
- What is Nano Banana 2.1 best used for in prompts?
- It's strongest on things that need exact, readable text or a steady character: infographics, signs, posters and multi-image follow-up chains. For loose or painterly styles, a model like Seedream 5 or Qwen Image 3 often fits better.
- What aspect ratio should I use for a panorama?
- Pick 4:1 for a true banner-style wide shot, and use the 4K resolution setting since the extra width needs the extra detail.
- How do I keep text spelled correctly in Nano Banana 2.1 images?
- Put the exact words in quotation marks in your prompt rather than describing them loosely, and if a word comes out wrong, correct it with a short follow-up instead of rewriting the whole prompt.
- Can Nano Banana 2.1 combine two separate photos into one image?
- Yes — open the Edit tool, attach both photos, and describe how they should combine in one sentence, for example placing a product from one photo into the scene from the other.
- How much does a Nano Banana 2.1 image cost in Plurel?
- Every Nano Banana 2.1 generation or edit costs 4 credits, regardless of aspect ratio or resolution.
- How do I keep the same character across several images?
- Generate the character once, then reply in the same chat with new scenes instead of writing fresh prompts — Plurel's chat edits whatever was just made, so the face and outfit carry over.
Keep going

Negative Prompt Guide: What to Leave Out of AI Images
What a negative prompt is, which AI image models still honor one, and the exact wording that avoids blurry hands, watermarks, and extra limbs.
Updated Oct 12, 2026

Grok Imagine Prompts: 10 That Work for Images and Clips
Grok Imagine prompts that work: ten copy-paste prompts for images and video, what the model responds to, where it wins, and the real credit cost.
Updated Oct 6, 2026

Consistent Character Prompt: Keep One Face Across Images
Learn how to write a consistent character prompt that holds one face across images and video, with anchor descriptions and reference images in Plurel.
Updated Sep 29, 2026

GPT Image 2.6 vs GPT Image 2.5: Which Should You Use?
GPT Image 2.6 vs GPT Image 2.5 compared: what changed, max resolution, credit price, and why 2.5 is no longer offered on Plurel.
Updated Nov 16, 2026