NanoBanana Image Generation Deployment Guide

Call Gemini's native image generation API through IKunCode

Resource	URL
IKunCode platform	api.ikuncode.cc

📋 Overview

NanoBanana is the Gemini image generation model family offered by IKunCode. You can call it directly with an IKunCode API Key — no VPN, no Google account required.

🎯 Model Overview

IKunCode offers two NanoBanana models:

Model	Model ID	Features	Use Cases
NanoBananaPro	gemini-3-pro-image-preview	Best quality, supports high-resolution output	High-quality posters, commercial assets, detailed creative work
NanoBanana2	gemini-3.1-flash-image-preview	Fast and affordable	Quick previews, batch generation, everyday use

💡 How to choose

• For the best image quality: choose NanoBananaPro (Pro version)
• For speed and cost-effectiveness: choose NanoBanana2 (Flash version)

📐 Supported Aspect Ratios

Both models support the following 10 aspect ratios:

Aspect Ratio	Description	Aspect Ratio	Description
1:1	Square image	3:2	Common camera ratio (landscape)
16:9	Standard landscape	2:3	Common camera ratio (portrait)
9:16	Standard portrait	21:9	Ultrawide
4:3	Classic landscape	5:4	Monitor ratio (landscape)
3:4	Classic portrait	4:5	Monitor ratio (portrait)

📏 Supported Resolutions

Each aspect ratio supports three resolution tiers:

1K Resolution (Quick Preview)

Aspect Ratio	Resolution	Aspect Ratio	Resolution
1:1	1024×1024	3:2	1232×816
16:9	1376×768	2:3	816×1232
9:16	768×1376	21:9	1584×672
4:3	1200×896	5:4	1136×896
3:4	896×1200	4:5	896×1136

2K Resolution (Recommended)

Aspect Ratio	Resolution	Aspect Ratio	Resolution
1:1	2048×2048	3:2	2464×1632
16:9	2752×1536	2:3	1632×2464
9:16	1536×2752	21:9	3168×1344
4:3	2400×1792	5:4	2272×1792
3:4	1792×2400	4:5	1792×2272

4K Resolution (Ultra HD)

Aspect Ratio	Resolution	Aspect Ratio	Resolution
1:1	4096×4096	3:2	4928×3264
16:9	5504×3072	2:3	3264×4928
9:16	3072×5504	21:9	6336×2688
4:3	4800×3584	5:4	4544×3584
3:4	3584×4800	4:5	3584×4544

🔧 API Request Format

NanoBanana uses the Google native format API, which differs from the OpenAI format.

Basic Request Structure

{
  "contents": [{
    "parts": [
      { "text": "Your image description" }
    ]
  }],
  "generationConfig": {
    "responseModalities": ["IMAGE"],
    "imageConfig": {
      "aspectRatio": "16:9",
      "image_size": "2K"
    }
  }
}

Parameters

Parameter	Description	Allowed Values
responseModalities	Response type, must be image	["IMAGE"]
aspectRatio	Aspect ratio	1:1, 16:9, 9:16, 4:3, 3:4, 3:2, 2:3, 21:9, 5:4, 4:5
image_size	Resolution tier	1K, 2K, 4K

API Endpoint

POST https://api.ikuncode.cc/v1beta/models/{ModelID}:generateContent

• NanoBananaPro: gemini-3-pro-image-preview
• NanoBanana2: gemini-3.1-flash-image-preview

📝 Sample Code

cURL Example

curl -X POST "https://api.ikuncode.cc/v1beta/models/gemini-3-pro-image-preview:generateContent" \
  -H "Authorization: Bearer sk-your-IKunCode-key" \
  -H "Content-Type: application/json" \
  -d '{
    "contents": [{
      "parts": [
        {"text": "A cute kitten sitting in a garden, oil painting style, HD, rich detail"}
      ]
    }],
    "generationConfig": {
      "responseModalities": ["IMAGE"],
      "imageConfig": {
        "aspectRatio": "16:9",
        "image_size": "2K"
      }
    }
  }'

Python Example

import requests
import base64

API_KEY = "sk-your-IKunCode-key"
API_URL = "https://api.ikuncode.cc/v1beta/models/gemini-3-pro-image-preview:generateContent"

payload = {
    "contents": [{
        "parts": [
            {"text": "A cute kitten sitting in a garden, oil painting style, HD, rich detail"}
        ]
    }],
    "generationConfig": {
        "responseModalities": ["IMAGE"],
        "imageConfig": {
            "aspectRatio": "16:9",
            "image_size": "2K"
        }
    }
}

headers = {
    "Authorization": f"Bearer {API_KEY}",
    "Content-Type": "application/json"
}

response = requests.post(API_URL, headers=headers, json=payload, timeout=600)

if response.status_code == 200:
    data = response.json()
    image_base64 = data["candidates"][0]["content"]["parts"][0]["inlineData"]["data"]

    # Save the image
    with open("output.png", "wb") as f:
        f.write(base64.b64decode(image_base64))
    print("Image saved to output.png")
else:
    print(f"Request failed: {response.status_code}")
    print(response.text)

Node.js Example

const fs = require('fs');

const API_KEY = 'sk-your-IKunCode-key';
const API_URL = 'https://api.ikuncode.cc/v1beta/models/gemini-3-pro-image-preview:generateContent';

async function generateImage() {
  const response = await fetch(API_URL, {
    method: 'POST',
    headers: {
      'Authorization': `Bearer ${API_KEY}`,
      'Content-Type': 'application/json'
    },
    body: JSON.stringify({
      contents: [{
        parts: [
          { text: 'A cute kitten sitting in a garden, oil painting style, HD, rich detail' }
        ]
      }],
      generationConfig: {
        responseModalities: ['IMAGE'],
        imageConfig: {
          aspectRatio: '16:9',
          image_size: '2K'
        }
      }
    })
  });

  const data = await response.json();
  const imageBase64 = data.candidates[0].content.parts[0].inlineData.data;

  fs.writeFileSync('output.png', Buffer.from(imageBase64, 'base64'));
  console.log('Image saved to output.png');
}

generateImage().catch(console.error);

🖼️ Image-to-Image Editing

NanoBananaPro also supports image-to-image editing. Upload an image along with an editing instruction to modify it.

{
  "contents": [{
    "parts": [
      {
        "inlineData": {
          "mimeType": "image/png",
          "data": "Base64-encoded string of the image"
        }
      },
      { "text": "Change the background to a starry sky, keep the person unchanged" }
    ]
  }],
  "generationConfig": {
    "responseModalities": ["IMAGE"],
    "imageConfig": {
      "aspectRatio": "16:9",
      "image_size": "2K"
    }
  }
}

💡 Tips for image-to-image editing

• Make editing instructions specific and clear — spell out what to keep and what to change
• Supports background swaps, style changes, adding elements, watermark removal, and more
• Encode the image as Base64 and place it in the inlineData.data field

🤖 Using It Inside Claude Code

If you have installed the ikunimage Skill, you can generate images with natural language directly in a Claude Code conversation — no manual API calls needed.

> Use ikun to draw a cyberpunk city nightscape, 16:9 ratio, 2K resolution

> ikun generate an ink-wash landscape painting, portrait 9:16

> Use ikun to edit this image /path/to/image.png and change the background to a sunset by the sea

For detailed installation and usage instructions, see: ikunimage AI Image Generation

⏱️ Performance Tips

Recommended Timeouts

Processing time varies significantly across resolutions — set a reasonable timeout:

Resolution	Recommended Timeout	Use Case
1K	360 seconds (6 minutes)	Quick previews, testing
2K	600 seconds (10 minutes)	Everyday use (recommended)
4K	1200 seconds (20 minutes)	Ultra HD output, commercial use

Bandwidth Considerations

Image data is transmitted as Base64, which results in large payloads:

• Use a stable, high-speed network connection
• The Base64 payload of a 4K image can exceed 10MB
• Avoid generating 4K images during peak network hours

⚠️ Notes

1. API Key: create an API Key with image model support on the IKunCode platform
2. API format: NanoBanana uses the Google native format, not OpenAI-compatible — keep this in mind
3. Resolution vs. time: higher resolutions take longer to generate; choose based on your real needs
4. Text rendering: rendering Chinese text inside images is supported (e.g. signs, poster copy) — just include it in your prompt
5. Pricing: prices vary by model and resolution; see the IKunCode pricing page for details