Google · Photo

Nano Banana Text-to-Image

Images from text

Google Nano Banana is a cutting-edge text-to-image model that generates images from natural language prompts.

from€0.08per render

Use in Genlaxy →

Examples

Nano Banana Text-to-Image: A little girl blowing soap bubbles in a backyard, sunlight making rainbow colors in the bubbles, candid photog — Prompt
A little girl blowing soap bubbles in a backyard, sunlight making rainbow colors in the bubbles, candid photography style.
Try this prompt →

Nano Banana Text-to-Image: A bustling New York street at sunset, golden light reflecting on skyscrapers, cinematic lighting — Prompt
A bustling New York street at sunset, golden light reflecting on skyscrapers, cinematic lighting
Try this prompt →

Nano Banana Text-to-Image: $28 — Prompt
$28
Try this prompt →

Nano Banana Text-to-Image: A breathtaking, National Geographic style photograph of a rustic stone farmhouse in Tuscany, Italy, at sunrise — Prompt
A breathtaking, National Geographic style photograph of a rustic stone farmhouse in Tuscany, Italy, at sunrise. A gentle morning mist blankets the rolling hills, partially obscuring the cypress trees that line a winding dirt road leading to the house. Warm, golden light from the rising sun breaks through the fog, illuminating the terracotta roof and a small wisp of smoke rising from the chimney. The scene feels timeless and tranquil. Shot on Kodak Portra 400 film for rich colors and soft grain. Wide-angle lens, capturing the grand scale and serene beauty of the landscape.
Try this prompt →

Specs

Typetext-to-image

OutputPhoto

InputText

Model IDgoogle/nano-banana/text-to-image

Similar models

GPT Image 2.5 Flare Text-to-ImageOpenAI€0.05 per render GPT Image 2.5 Sunburst Text-to-ImageOpenAI€0.05 per render Seedream V5.0 ProBytedance€0.09 per render Seedream V5.0 FlashBytedance€0.06 per render Qwen Image 3.0 Text-to-ImageAlibaba€0.06 per render Qwen Image 3.0 Pro Text-to-ImageAlibaba€0.08 per render

All models →

Prices are per render and charged from your balance only when a render succeeds. Failed renders are refunded automatically. Examples show what the model can do; your result depends on your prompt and settings.