Nano Banana API · Ace Data Cloud

Nano Banana API:
Generate AI Images

Powered by Google Gemini 2.5 Flash/3 Pro Image models, enabling high-quality image generation and editing through a simple REST API. Supports multiple scenarios including text-to-image, multi-image editing, virtual try-on, and product background replacement.

🖼️ 2 Operation Modes 🤖 2 Model Versions 📐 7 Aspect Ratios 🔑 Bearer Token Authentication
🖼️
2
Operation Modes
🤖
2
Model Versions
📡
2
API Endpoints
📐
7
Aspect Ratio Options

What Is Nano Banana?

Nano Banana is a unified API wrapper based on Google Gemini image generation models. It simplifies and optimizes the Gemini 2.5 Flash Image and Gemini 3 Pro Image models, providing a more user-friendly interface and lower pricing.

Ace Data Cloud offerslower prices than the official service (approximately 1/3 of the official price), along withhigher concurrency support, dedicated customer service, and 24/7 stability assurance. No need to handle complex Google Cloud configurations—ready to use out of the box.

Core Capabilities of the Nano Banana API

Text-to-image generation and multi-image editing, covering the complete image creation workflow

✨

High-Quality Text-to-Image Generation

Enter detailed text descriptions (subject, mood, lighting, camera) to generate professional-grade, high-quality images. Suitable for brand visuals, product posters, cover images, and more.

🔀

Multi-Image Combined Editing

Upload multiple images (such as a portrait + clothing), and AI will intelligently blend them. Supports advanced editing operations such as virtual try-on, product placement, and scene compositing.

👗

Virtual Outfit Change

Provide a portrait photo and a clothing PNG to automatically create a dressed-up result. Ideal for e-commerce apparel displays and quickly expanding livestream image content.

🏠

Product Background Replacement

Place white-background product images into real scenes (wooden tables, grass, bathrooms, desks) to make product detail pages richer and more lifelike.

🎨

Attribute Replacement

Change materials, colors, and styles. Swap a bag to leather, a drink to another flavor, or shoes to a different colorway—making visual SKU experimentation easier.

⚡

Asynchronous Webhook

Set callback_url, and results will be automatically pushed after the task is completed. No polling required, ideal for batch production scenarios.

AI-Generated Work Showcase

The images below were actually generated by the Nano Banana API—what you see is what you get

cURL
curl -X POST https://api.acedata.cloud/nano-banana/images \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "action": "generate",
    "prompt": "白瓷咖啡杯,晨光窗台,85mm镜头,浅景深,柔和高光,治愈干净",
    "model": "nano-banana",
    "aspect_ratio": "3:2"
  }'
Response
{
  "success": true,
  "task_id": "c1075552-a527-4a70-8d98-ef783f4087c1",
  "trace_id": "3352b744-d17e-4d01-b1ef-a8c8e2ca3d2e",
  "data": [{
    "image_url": "https://cdn.acedata.cloud/d635720766.png"
  }]
}

Quick Start—Get Started in 3 Minutes

A simple REST API with Bearer Token authentication. Generate your first AI image with a single request.

1

Get an API Key

Register with Ace Data Cloud and get your Bearer Token from the console

2

Send a POST Request

Send a request to /nano-banana/images with a prompt and parameters

3

Get the Image URL

Receive a permanently accessible image URL—ready for production use immediately

Use Cases

From e-commerce content to brand creativity—Nano Banana covers these scenarios

👗

E-commerce Apparel

Have an influencer take a single styled photo, then use different clothing PNGs to batch-generate outfit images and quickly expand your store assets

📦

Product Detail Pages

Use a single product image on a white background to compose images in multiple settings such as wooden tables, grass, and bathrooms, making detail pages richer and more convincing

🎨

Brand Visuals

Quickly generate brand key visuals, campaign posters, and cover images. Switch the same key visual between styles and holiday themes with one click

🤖

AI Agent Integration

Integrate with Claude and ChatGPT through an MCP Server or Dify plugin to enable natural-language image generation

Advanced Features

Nano Banana is more than simple text-to-image generation—it can also handle these advanced tasks

🎭

2D to 3D Figurines

Convert anime characters/portraits into commercialized figurine models, complete with packaging box displays

🔧

Product Exploded Views

Showcase a product's internal structure, with metal parts and electronic components separated and suspended for a technical aesthetic

📷

Old Photo Restoration

Restore scratched and damaged old black-and-white photos, enhance clarity, and colorize them naturally

🚗

Multi-Element Composites

Naturally blend multiple assets such as models, products, and backgrounds into a single high-quality image

+ Multi-angle shots + Meme generation + Sketch to photorealistic image + Map to real-world scene

Get Started in 3 Quick Steps

From registration to generating your first AI image, it takes less than 3 minutes

01

Register and Get an API Key

Create a free account on Ace Data Cloud. Generate your Bearer Token from the API management console.

02

Make Your First API Call

Send a POST request with a text prompt to /nano-banana/images. You can use an SDK, cURL, or any HTTP client.

03

Integrate and Scale

Embed the API into your application. Use Webhooks for asynchronous processing and scale confidently to production.

Why Choose Ace Data Cloud?

Compare with Google Official and see our advantages

Feature Ace Data Cloud Google Official
Price As low as $0.0133 / request $0.039 Interactive / $0.02 Batch
Setup Requirements ✓ Ready to use instantly ✗ Google Cloud configuration required
Concurrency Support ✓ High-concurrency guarantee ✗ Low default quota
Webhook Callbacks ✓ Partially supported
Task Polling ✓ Free —
MCP Server ✓ ✗
Dify Plugin ✓ ✗
Dedicated Customer Support ✓ ✗
24/7 Monitoring ✓ Partially supported

Which model is right for you?

Choose the right model based on your quality and speed needs

Recommended

nano-banana

Fast

Based on Gemini 2.5 Flash Image. Faster generation speeds and lower costs, ideal for batch content production and prototype validation.

  • ✓ Faster response times
  • ✓ As low as $0.0133 / call
  • ✓ 7 aspect ratios
  • ✓ Ideal for high-frequency scenarios

nano-banana-2-lite

Lightweight

Based on Gemini 3.1 Flash Lite Image. Supports 1K resolution only, with fast image generation in 3-5 seconds.

  • ✓ Gemini 3.1 lightweight model
  • ✓ As low as $0.0133 / call
  • ✓ Supports 1K resolution only
  • ✓ Ideal for rapid validation
New

nano-banana-2

Balanced

Based on Gemini 3.1 Flash Image. Pro-level quality + Flash speed, a great value choice.

  • ✓ Pro-level image quality
  • ✓ Flash-level response speed
  • ✓ As low as $0.0267 / call
  • ✓ Ideal for most scenarios

nano-banana-pro

High Quality

Based on Gemini 3 Pro Image. Highest-quality output with support for 2K/4K resolution. Ideal for flagship products and demanding scenarios.

  • ✓ Highest image quality
  • ✓ As low as $0.0333 / call
  • ✓ Supports 1K/2K/4K resolution
  • ✓ Ideal for brand visuals

Nano Banana API Pricing

Transparent pay-as-you-go pricing. No subscription fees. No hidden costs. Pay only for what you use.

Pay as you go
Standard
$0.0133 / call

nano-banana model (Gemini 2.5 Flash Image)
Save up to 27% with volume packages

  • ✓ Text-to-image (Generate)
  • ✓ Multi-image editing (Edit)
  • ✓ 7 aspect ratios
  • ✓ Webhook callbacks
  • ✓ Task polling—Free
  • ✓ nano-banana-2-lite: as low as $0.0133 / call
  • ✓ nano-banana-2: as low as $0.0267 / call
  • ✓ nano-banana-pro: as low as $0.0333 / call
View pricing details View API documentation
Enterprise
Custom

Volume discounts for high-usage teams

  • ✓ Usage-based discounts
  • ✓ Priority support
  • ✓ Dedicated account manager
  • ✓ Custom rate limits
  • ✓ SLA guarantee
Contact sales

Frequently Asked Questions

Everything you need to know about using the Nano Banana API

What model is Nano Banana? ▾

Nano Banana is an alias wrapper for Google Gemini image generation models. nano-banana corresponds to Gemini 2.5 Flash Image, nano-banana-2-lite corresponds to Gemini 3.1 Flash Lite Image (1K only, fast image generation), nano-banana-2 corresponds to Gemini 3.1 Flash Image (Pro-level quality + Flash speed), and nano-banana-pro corresponds to Gemini 3 Pro Image. We provide a simpler interface and lower prices.

How does pricing work? ▾

Pay as you go, with no subscription fee. nano-banana and nano-banana-2-lite models start at $0.0133/request, nano-banana-2 models start at $0.0267/request, and nano-banana-pro models start at $0.0333/request. Task status polling is always free. Bulk packages offer discounts of up to 27%.

What image operations are supported? ▾

Two operation modes: Generate (text-to-image) and Edit (multi-image editing). Generate supports 7 aspect ratios (1:1, 3:2, 2:3, 16:9, 9:16, 4:3, 3:4), and Pro models also support 1K/2K/4K resolution options. Edit mode supports uploading up to 10 images for combined editing.

How should I handle longer generation times? ▾

Two methods: (1) Use callback_url—set a Webhook URL, and we will POST the result when complete. (2) Poll via /nano-banana/tasks—this endpoint is free, and you can check the status at any time.

Are image URLs permanently valid? ▾

Yes. The returned image_url is stored on our CDN and is permanently accessible, so it can be used directly in production environments.

Can it be used with Claude or ChatGPT? ▾

Yes! We provide an MCP Server (mcp-nano-banana) and a Dify plugin, which can integrate directly with Claude Desktop, VS Code, and Dify workflows. Your AI Agent can generate and edit images through natural language.

How do I write a good Prompt? ▾

Recommended structure: subject + atmosphere + lighting + lens + texture keywords. For example: "white porcelain coffee cup, morning-light windowsill, 85mm portrait lens, shallow depth of field, soft highlights, soothing, clean, relaxed." The more specific the description, the better the generated result.

What should I keep in mind for Edit mode? ▾

Edit mode requires passing 1-10 images through the image_urls parameter (HTTP/HTTPS URL or Base64). Each image can be up to 10MB. The Prompt needs to describe how to process these images, for example, "make this person wear this T-shirt."

Start Generating AI Images Now

As low as $0.0133 per call—lower prices than the official service, a simpler interface, and higher concurrency support. Pay as you go—no subscription fees, no commitments.