Přejít na obsah
Přihlásit seZačněte zdarma

How to write prompts for Nano Banana 2.1

Nano Banana 2.1 makes pictures from a sentence and edits the pictures you give it. This guide rests on Google’s own documentation, put in our words. We have made only one sample picture with this model, on its model page, which is too little to test from, so this guide states nothing as our test.

Written from the provider’s documentation alone, with no result of ours to quote. Last checked November 8, 2026.

Make something with Nano Banana 2.1The Nano Banana 2.1 model page

Be specific, and say what the picture is for

From the provider’s documentation

  • Google’s documentation asks for exact materials, patterns and shapes instead of a vague label: say what a coat is made of and what colour and cut it has, rather than calling it a nice coat.
  • It says that stating the purpose helps, for example that the picture is a logo for a skincare brand or the cover of a menu, because the model then makes better choices.
  • It advises describing what you want rather than what you do not want: ask for an empty street, not a street without cars.
  • For a complex scene it suggests building in order: the background first, then the things in front, then the main subject. Camera words such as wide-angle, macro or low angle steer the framing.

Text, edits and several pictures

From the provider’s documentation

  • For text in the picture, Google says to put the exact words in quotes, describe the lettering in plain words such as clean, bold and sans-serif, and state the layout and colours. Its list of limits adds a tip: write the text first, then ask for the picture that contains it.
  • Treat the first result as a draft and follow with small, targeted changes, such as warmer light or a more serious face, naming what must stay as it is.
  • To edit a picture, give it with a short instruction. To change one thing, name that thing and say that everything else stays. To combine several pictures, call them the first, the second and so on, and describe the finished scene, including how light and shadow should match.
  • Google says Nano Banana 2.1 keeps the look of up to four characters and the detail of up to ten objects steady within one piece of work.

Limits and positioning Google lists

From the provider’s documentation

  • It does not always return the exact number of pictures you asked for.
  • Google names the languages it does well in, including English, German, Spanish, French, Japanese, Korean, Portuguese and Simplified Chinese.
  • Google presents Nano Banana 2.1 as its main model for new work, and Nano Banana Pro for complex tasks that need heavier reasoning and exact text.

Settings of this model

Od
Google
Typ
AI image model
Způsob tvorby
Text k obrázku · Úprava obrázku
Rozlišení
1K / 2K / 4K
Cena
from 58 / image

Otázky

Does Nano Banana 2.1 follow prompts in Korean or Japanese?

Google names Korean and Japanese among the languages it does well in. We have not tested this ourselves, so check the first result before you rely on it.

Should I write what I do not want in the picture?

Google’s documentation advises against it. Describe the scene you want instead, for example an empty street instead of a street with no traffic.

How do I get exact words into the picture?

Put the words in quotes, describe the lettering and layout, and check the spelling. Google also suggests writing the text first and then asking for the picture that contains it. We have no test of our own on this.

Provider documentation we read

We read these pages and wrote the points that matter in our own words. The provider may change them.

More prompting guides

Názvy modelů, společností a produktů na tomto webu jsou ochrannými známkami jejich vlastníků; Pickshot s nimi není spojen, sponzorován ani jimi schválen.