Synthome Docs
Models

Nano Banana Pro

Google's Gemini 3 Pro image generation with advanced typography

Nano Banana Pro

Google's Gemini 3 Pro Image model with advanced text rendering and reasoning capabilities.

PropertyValue
Model IDgoogle/nano-banana-pro
ProvidersReplicate, Fal
TypeImage generation

Basic Usage

import { compose, generateImage, imageModel } from "@synthome/sdk";

const execution = await compose(
  generateImage({
    model: imageModel("google/nano-banana-pro", "replicate"),
    prompt:
      "A book cover with the title 'Adventures in AI' in elegant typography",
  }),
).execute();

Key Features

  • Accurate text rendering: Generates readable text in images
  • Advanced reasoning: Better understanding of complex prompts
  • High resolution: Superior detail and quality
  • Multi-image support: Reference multiple images

Typography & Text

Nano Banana Pro excels at generating images with text:

generateImage({
  model: imageModel("google/nano-banana-pro", "replicate"),
  prompt: "A vintage poster with 'SUMMER SALE' in bold retro lettering",
});

Marketing Materials

generateImage({
  model: imageModel("google/nano-banana-pro", "replicate"),
  prompt:
    "Instagram post design with text 'New Collection' in modern sans-serif font, fashion photography background",
  aspectRatio: "1:1",
});

Logo Concepts

generateImage({
  model: imageModel("google/nano-banana-pro", "replicate"),
  prompt:
    "Minimalist logo design with the letters 'AB' intertwined, black on white",
});

Options

OptionTypeDefaultDescription
promptstringrequiredText description
imagestring | string[]-Input image(s) for transformation
aspectRatiostring1:1Output aspect ratio
outputFormat"jpg" | "png" | "webp"pngOutput format

Complex Scene Generation

Better at understanding and rendering complex scenes:

generateImage({
  model: imageModel("google/nano-banana-pro", "replicate"),
  prompt:
    "A busy Tokyo street at night with neon signs in Japanese, rain reflections on the ground, crowds of people with umbrellas",
  aspectRatio: "16:9",
});

Multi-Reference Generation

Combine multiple reference images:

generateImage({
  model: imageModel("google/nano-banana-pro", "replicate"),
  prompt:
    "Create a new design combining the color palette of the first image with the style of the second",
  image: ["https://example.com/colors.jpg", "https://example.com/style.jpg"],
});

Examples

Book Cover

generateImage({
  model: imageModel("google/nano-banana-pro", "replicate"),
  prompt:
    "Science fiction book cover, title 'NEXUS' in chrome metallic text, futuristic city background",
  aspectRatio: "2:3",
});

Event Flyer

generateImage({
  model: imageModel("google/nano-banana-pro", "replicate"),
  prompt:
    "Concert flyer with 'LIVE MUSIC FRIDAY' headline, acoustic guitar illustration, warm earth tones",
  aspectRatio: "4:5",
});

UI Mockup

generateImage({
  model: imageModel("google/nano-banana-pro", "replicate"),
  prompt:
    "Mobile app login screen mockup with 'Welcome Back' heading, clean minimal design, light mode",
  aspectRatio: "9:16",
});

When to Use Pro

Choose Nano Banana Pro over standard Nano Banana when:

  • Text must be readable and accurate
  • Creating marketing materials with typography
  • Complex multi-element scenes
  • Higher quality output is needed
  • You need better prompt understanding

For simple image generation without text, standard Nano Banana is faster and more cost-effective.

How is this guide?