Nano Banana Prompts: 600+ Copy-Paste Examples
600+ Nano Banana prompts from the Hooked library, plus a guide to Google's prompt formula for text-to-image and reference-based edits. "Nano Banana" now usually means Nano Banana 2 (Gemini 3.1 Flash Image); the original was Gemini 2.5 Flash Image. The library was written for Nano Banana Pro and works on Nano Banana 2 too.
★★★★★ 4.9/5 · 27,000+ users
Updated September 25, 2026.
Prompt structure
The Nano Banana prompt formula
Google's prompting guide (Google Cloud blog, March 2026) gives a five-part text-to-image formula and a separate one for references: say what each image is for ('use the sketch as structure and the fabric as texture') and then describe the new scene. Because the model reasons before it renders, full sentences that describe a scene work better than keyword lists. Editing is conversational: say what to change and, explicitly, what to keep.
Strengths and limits
What Nano Banana is good at
Good at
- Fast generation and conversational editing: change one thing in a follow-up turn and keep the rest
- Combining up to 14 reference images in one generation
- Real-time Google Search grounding, so it can draw current subjects, weather or data
- Readable text in images and translating it into other languages, for posters, labels and ads
- Very wide or tall formats: 1:4, 4:1, 1:8 and 8:1 on top of the usual ratios
Limits
- Resolutions: 512px, 1K, 2K and 4K on Nano Banana 2; Flash Lite is limited to 1K
- Aspect ratios: 1:1, 3:2, 2:3, 3:4, 4:3, 4:5, 5:4, 9:16, 16:9, 21:9, plus 1:4, 4:1, 1:8, 8:1 on Nano Banana 2
- Knowledge cutoff January 2025 for the model itself; use search grounding for anything newer
- Every generated image carries an invisible SynthID watermark
- Output formats JPEG and PNG
From prompt to posted video
Paste the prompt, get the video, post it
Hooked Pro is $29 a month for 1,000 credits across Veo 3.1, Kling 3.0, Seedance 2.0, Nano Banana Pro and GPT Image 2, plus AI UGC ads, captions and TikTok slideshows in the same plan. Or pay $89 once for the Studio and run it on your own API keys, at the provider's price.
Better results
Tips that make a difference
Do
- Describe the scene in sentences instead of listing keywords; the guide lists 'generic keyword lists' among things to avoid.
- Phrase exclusions positively: 'an empty street', not 'no cars'.
- Put any text you want rendered in quotes and name the font style: 'the headline "SUMMER DROP" in bold white sans-serif'.
- Direct like a photographer: name the lighting set-up ('three-point softbox', 'golden-hour backlight'), the lens ('f/1.8 shallow depth of field', 'macro') and the camera feel ('disposable camera', 'GoPro').
- Name materials, not categories: 'navy tweed blazer' rather than 'suit jacket', 'matte ceramic mug' rather than 'mug'.
- When editing, state what must stay unchanged (face, product label, background) along with the change.
- With several references, say which image supplies what: the character, the product, the style.
- For time-sensitive visuals (weather, scores, events), ask it to search first and then say how to turn the result into an image.
- Iterate in small follow-up turns rather than rewriting the whole prompt.
Common mistakes
- Keyword soup ('4k, ultra detailed, trending, masterpiece') instead of a described scene
- Negative phrasing ('no people', 'without text') instead of describing what should be there
- Editing without saying what to preserve, so faces or labels change
- Uploading several references without assigning a role to each one
- Expecting current facts from the model without turning on search grounding
Tested prompts
634 Nano Banana prompts from the Hooked library
Each prompt was run and kept with its result. Open one to see the full text and the output.

YouTube Thumbnail - Cinematic Close-Up of Crispy Chicken Nuggets
A slow, cinematic close-up video of crispy golden chicken nuggets. One nugget is bitten in half, revealing tender, juicy shredded white meat inside. The extra-crunchy panko coating glistens with golden-brown highlights and coarse sea salt f…
Full prompt and result
YouTube Thumbnail - Cinematic Talk Show Portrait (Emilia Clarke Inspired)
A photorealistic, cinematic talk-show portrait of a beautiful actress with rich chestnut-brown hair styled in soft, glossy waves, parted slightly to the side, and a bright, warm, infectious smile that radiates charm and friendliness, {argum…
Full prompt and result
YouTube Thumbnail - Cosmic Chessboard Checkmate Scene
“Eternity’s Checkmate, a massive cosmic chessboard floating in deep outer space, final move frozen in time, a radiant cracked king chess piece emitting starlight energy, a mysterious hooded cosmic entity formed from galaxies and nebula dust…
Full prompt and result
YouTube Thumbnail - Hyper-Detailed Hamster Weightlifting Cinematic Scene
A hyper-detailed, adorable golden hamster in a bright, cheerful gym, animated in smooth cinematic motion. The chubby hamster, fluffy and round, lifts oversized black dumbbells with both tiny paws, bouncing slightly with effort. It wears pla…
Full prompt and result
YouTube Thumbnail - K-pop Star Jennie Kim Stage Performance Portrait
{ "subject": { "name": "Jennie Kim", "alias": "Jennie", "group": "BLACKPINK", "persona": "Elegant, fierce stage presence, signature 'human Chanel' style", "expression": "Focused, slightly intense, sharp gaze off-camera…
Full prompt and result
YouTube Thumbnail - Kollywood Cinematic Film Post Photo Prompt
Kollywood cinematic film Post photo prompt Gemini Nano Banana creation.
Full prompt and result
YouTube Thumbnail - Panning shot of a soccer player kicking a ball
Panning shot of a mysterious soccer player in {argument name="jersey color" default="red"} professional soccer jersey mid air kicking a ball, motion blur, cinematic lighting against a Light green gradient background with streaking lights --…
Full prompt and result
YouTube Thumbnail - Surreal Campfire in Cosmic Void
{ "type": "image_generation_prompt", "style": "surreal, cinematic, conceptual art", "camera": { "angle": "high-angle", "shot": "full shot", "composition": "symmetrical balance with central campfire dividing two characters"…
Full prompt and result
YouTube Thumbnail - Tilt-Shift Tokyo Diorama at Sunset
Tilt-shift 3D model diorama in 4:5 aspect ratio, highlighting {argument name="landmark" default="Tokyo Tower and the Shibuya Crossing"} at {argument name="time of day" default="sunset"}. Tiny pedestrians and cyclists navigate the bustling s…
Full prompt and result
YouTube Thumbnail - Vlogger Thumbnail Image Generation Prompt
Generate an illustration for a thumbnail for late-night relaxed piano practice
Full prompt and result
FAQ
Frequently asked questions
Which model is Nano Banana?
Today the name usually refers to Nano Banana 2, Google's Gemini 3.1 Flash Image, launched on February 26, 2026. The original Nano Banana was Gemini 2.5 Flash Image, now a legacy model.
What is the difference between Nano Banana 2 and Nano Banana Pro?
Nano Banana 2 is the faster and cheaper model for quick generation and edits. Nano Banana Pro (Gemini 3 Pro Image) is meant for images that have to be correct, such as infographics, diagrams and designs with text.
Do these prompts work on Nano Banana 2?
Yes. The library was written for Nano Banana Pro, and the prompts also run on Nano Banana 2.
How do I edit an image without changing the face or product?
Say what to change and, in the same prompt, what must stay the same, such as the face, the label or the background. Make changes in small follow-up turns.
Does Nano Banana add a watermark?
Every image carries an invisible SynthID watermark from Google.
Keep exploring
Related prompt libraries and pricing
Sources
- https://cloud.google.com/blog/products/ai-machine-learning/ultimate-prompting-guide-for-nano-banana
- https://ai.google.dev/gemini-api/docs/image-generation
- https://blog.google/innovation-and-ai/technology/ai/nano-banana-2/
- https://blog.google/innovation-and-ai/technology/developers-tools/build-with-nano-banana-2/
- https://deepmind.google/models/gemini-image/flash/
Two ways in · pick yours
Every template here opens in the app — 1,600+ and counting.
Pay monthly and we run the AI for you, credits included — or pay $89 once, own the app for life and bring your own keys.
Pro
Most flexibleWe run the AI for you. No API keys, nothing to set up — open the app and generate.
1,000 credits a month · cancel anytime
Secure checkout by Stripe · Cancel anytime
- About 10 UGC ads or 250 images a month
- Every video format: UGC, avatars, slideshows…
- 83 AI avatars & 76 voices in 25 languages
- Publish to TikTok, Instagram & YouTube
- 3 channels, 3 auto-channels, API included
Need more? One-time credit packs from $19. They never expire.
4.9/5 · 27,000+ users create with Hooked