Nano Banana Pro

Nano Banana Pro, or Gemini 3 Pro Image in Google's naming, came out in November 2025 and sits at the top of the Nano Banana family. Its main trick is combining many references in one frame: the person from one photo, the jacket from another, the room from a third. If you need a finished marketing visual rather than a quick sketch, this is the model to try.

What you can make

• Group shots assembled from separate photos of people or products.
• 4K key art and campaign visuals.
• A new subject rendered in the style of your mood board.
• Infographics and explainer pictures where text and imagery share the frame.

Settings that matter

• References: up to 14 per request. Point to them by number, like "the man from image 2 holding the cup from image 5".
• Resolution: 1K (default), 2K or 4K. Lock the composition at 1K and only then go up.
• Aspect ratio: 10 shapes from 1:1 to 21:9. Kadraris defaults to 1:1, so set the ratio yourself on edits.
• Prompt: up to 32,000 characters in English or Chinese. Full sentences work better than keyword lists, which is also what Google recommends for its image models.
• Output: one image per call; for variants, send several calls in parallel.

Starter prompt

The chef from image 1 stands behind the counter from image 2, plating the dessert from image 3. Warm evening light, shallow depth of field, editorial food magazine style, 4K

Renumber the roles to match the order in which you upload your own references.

Questions

What does Nano Banana Pro cost per image?

from $0.07 per image, deducted from a prepaid balance. The estimate is on screen before you submit, and failed images cost nothing.

Will I hit limits like in the Gemini app?

No allowance refills every few hours here. Each image is paid from a balance that does not expire; the Nano Banana and Veo limits guide explains the difference.

Is it a reasonable Midjourney replacement?

For many jobs, yes, and it comes with an official API and per-image billing. See Midjourney alternatives with an API, the Krea AI and Leonardo AI guides, and the image model ranking for 2026. For speed and four images per run, Nano Banana 2 is the lighter option.

Nano Banana Pro in your code

curl https://kadraris.com/v1/generate?wait=60 \
  -H "Authorization: Bearer $KADRARIS_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "model": "nano-banana-pro",
  "input": {
    "prompt": "Packshot of a white linen sneaker on a concrete block, hard midday sun, crisp shadow, 85mm",
    "aspect_ratio": "1:1",
    "resolution": "1k"
  }
}'

import os, time, requests

API = "https://kadraris.com/v1"
H = {"Authorization": f"Bearer {os.environ['KADRARIS_API_KEY']}"}

task = requests.post(f"{API}/generate", headers=H, json={
    "model": "nano-banana-pro",
    "input": {
        "prompt": "Packshot of a white linen sneaker on a concrete block, hard midday sun, crisp shadow, 85mm",
        "aspect_ratio": "1:1",
        "resolution": "1k"
    }
}).json()
while task["status"] in ("queued", "running"):
    time.sleep(3)
    task = requests.get(f"{API}/tasks/{task['id']}", headers=H).json()

print(task["status"], task["cost"], [o["url"] for o in task["output"]])

const API = "https://kadraris.com/v1";
const headers = { Authorization: `Bearer ${process.env.KADRARIS_API_KEY}`, "Content-Type": "application/json" };

let task = await (await fetch(`${API}/generate`, {
  method: "POST",
  headers,
  body: JSON.stringify({
    "model": "nano-banana-pro",
    "input": {
      "prompt": "Packshot of a white linen sneaker on a concrete block, hard midday sun, crisp shadow, 85mm",
      "aspect_ratio": "1:1",
      "resolution": "1k"
    }
  }),
})).json();

while (task.status === "queued" || task.status === "running") {
  await new Promise((r) => setTimeout(r, 3000));
  task = await (await fetch(`${API}/tasks/${task.id}`, { headers })).json();
}
console.log(task.status, task.cost, task.output.map((o) => o.url));

Input fields

Fetch this list as JSON with GET /v1/models/nano-banana-pro.