How to use nano banana pro 4K parameter？

[violetfantasty]

I can not find how to set 4K parameter for nano banana pro in openrouter docs.

[violetfantasty]

I tried to put image_size and resolution into extra_body and image_config, but it doesn’t work.

[cc-320-copy]

I tried to put image_size and resolution into extra_body and image_config, but it doesn’t work.

I ran into the same problem. Have you solved it? If you have, please let me know.

[aditya-pushkar]

Hi guys, i was facing the same issue with sdk, so I found an alternative solution using the API directly. Below is the code snippet for reference

const resolution = "2k" 

const response = await fetch("https://openrouter.ai/api/v1/chat/completions", {
      method: "POST",
      headers: {
        "Authorization": `Bearer ${process.env.OPENROUTER}`,
        "Content-Type": "application/json",
      },
      body: JSON.stringify({
        model: "google/gemini-3-pro-image-preview",
        messages: [{ role: 'user', content: contents }],
        modalities: ["text", "image"],
        image_config: {
          aspect_ratio: aspectRatio,
          image_size: resolution,
        },
        stream: false
      })
    }); 

[No5972]

OpenRouter HTTP API should support specifying generating image resolution now. Setting image_config.image_size directly can control the resolution of generated images. Of course it is for Gemini only at the moment. Their docs' having been updated can be seen.

(Code below is from updated docs of link above cited on December 23, 2025, and for reference only)

import requests
import json

url = "https://openrouter.ai/api/v1/chat/completions"
headers = {
    "Authorization": f"Bearer {API_KEY_REF}",
    "Content-Type": "application/json"
}

payload = {
    "model": "google/gemini-2.5-flash-image-preview",
    "messages": [
        {
            "role": "user",
            "content": "Create a picture of a nano banana dish in a fancy restaurant with a Gemini theme"
        }
    ],
    "modalities": ["image", "text"],
    "image_config": {
        "aspect_ratio": "16:9",
        "image_size": "4K"
    }
}

response = requests.post(url, headers=headers, json=payload)
result = response.json()

if result.get("choices"):
    message = result["choices"][0]["message"]
    if message.get("images"):
        for image in message["images"]:
            image_url = image["image_url"]["url"]
            print(f"Generated image: {image_url[:50]}...")

[zuojianghua]

Generating images at 2K resolution works without problems, but simply changing the parameters to 4K results in a very high probability of the API call failing. Could this be due to the transmitted file size being too large?