# Nano Banana Pro

Google’s best. Fourteen reference pictures in one edit, which is exactly what a character sheet needs.

Model id: `nano-banana-pro`
Type: Image
Made by: Google
Inputs: Text, References
Resolution: 1K, 2K, 4K
Aspect ratio: auto, 21:9, 16:9, 3:2, 4:3, 5:4, 1:1, 4:5, 3:4, 2:3, 9:16
Updated: Oct 12, 2026

## Pricing

| Setting | MiniAPI | fal | Difference |
| --- | ---: | ---: | ---: |
| 1K/2K | $0.165 / image | $0.15 | +10% |
| 4K | $0.33 / image | $0.30 | +10% |

You are not charged for a job that fails.

## Parameters

| Name | Required | Values | Default | Description |
| --- | --- | --- | --- | --- |
| `prompt` | Yes | text |  | What to make, in plain words. |
| `image_urls` | Yes | https URLs |  | Pictures to use as references. |
| `num_images` | No | 1 to 4 | 1 | How many images to make. Each one is charged. |
| `aspect_ratio` | No | auto, 21:9, 16:9, 3:2, 4:3, 5:4, 1:1, 4:5, 3:4, 2:3, 9:16 | auto | Shape of the result, for example 16:9 or 9:16. |
| `output_format` | No | jpeg, png, webp | png | File type of the result. |
| `safety_tolerance` | No | 1 to 6 | 4 | How strict the model's own content filter is. Lower is stricter. |
| `system_prompt` | No | text |  | Optional system instruction that steers the model's persona and output style across the request. Leave blank to omit; when provided, it is sent as the system instruction to Gemini (or as a system message on OpenAI-compatible providers). |
| `resolution` | No | 1K, 2K, 4K | 1K | Output size. Higher is sharper, slower and costs more. |
| `limit_generations` | No | true, false | true | Experimental parameter to limit the number of generations from each round of prompting to 1. Set to `True` to to disregard any instructions in the prompt regarding the number of images to generate. |
| `enable_web_search` | No | true, false | false | Enable web search for the image generation task. This will allow the model to use the latest information from the web to generate the image. |
| `seed` | No | integer |  | The same seed with the same settings gives a similar result. Leave it out for a random one. |

## Quickstart

```bash
curl -X POST "https://api.minicut.ai/v1/queue/nano-banana-pro" \
  -H "Authorization: Key $MINIAPI_KEY" \
  -H "Content-Type: application/json" \
  -d '{"prompt":"a red fox in fresh snow, morning light","image_urls":["https://example.com/ref-1.jpg"]}'
```

```json
{
  "request_id": "0192f3a4-5b6c-7d8e-9f01-23456789abcd",
  "status": "IN_QUEUE",
  "queue_position": 0,
  "status_url": "https://api.minicut.ai/v1/queue/nano-banana-pro/requests/0192f3a4-5b6c-7d8e-9f01-23456789abcd/status",
  "response_url": "https://api.minicut.ai/v1/queue/nano-banana-pro/requests/0192f3a4-5b6c-7d8e-9f01-23456789abcd",
  "cancel_url": "https://api.minicut.ai/v1/queue/nano-banana-pro/requests/0192f3a4-5b6c-7d8e-9f01-23456789abcd/cancel",
  "price": "0.165000",
  "currency": "USD"
}
```

Page: https://api.minicut.ai/models/nano-banana-pro
API reference: https://api.minicut.ai/docs/api