# Nano Banana Pro (Gemini 3 Pro Image)

> Nano Banana Pro (Gemini 3 Pro Image) on Clonvo: image generation, $2.00 per 1M input tokens and $120.00 per 1M image output tokens.

Google's most capable image generation and editing model, built on Gemini 3 Pro, with 1K, 2K or 4K output.

-   Model id: `google/gemini-3-pro-image`
-   Provider: Google AI Studio
-   Released: May 28, 2026
-   Image generation: up to 4 images per request, up to 14 reference images. See [Image generation](https://clonvo.com/docs/images.md).

## Pricing

The provider’s list price in USD. Requests on credits are charged exactly this; each response reports its cost in usage.cost.

| Charge | Price | Unit |
| --- | --- | --- |
| Input (prompt) | $2.00 | per 1M tokens |
| Output (completion) | $12.00 | per 1M tokens |
| Image output | $120.00 | per 1M tokens |

## Capabilities

-   Vision
-   Reasoning

### Supported parameters

Upstream parameters this model honours. Others may be ignored.

-   `prompt`
-   `n`
-   `resolution`
-   `aspect_ratio`
-   `size`
-   `seed`
-   `input_references`

## Quick start

Use the OpenAI SDK with Clonvo’s base URL and this model id.

```python
import os
from openai import OpenAI

client = OpenAI(
    base_url="https://api.clonvo.com/v1",
    api_key=os.environ["CLONVO_API_KEY"],
)

completion = client.chat.completions.create(
    model="google/gemini-3-pro-image",
    messages=[{"role": "user", "content": "Summarise the plot of Hamlet in one sentence."}],
)
print(completion.choices[0].message.content)
```

## Data policy

Requests to this model are sent to Google AI Studio, whose API terms govern how they handle your data: [Google AI Studio data policy](https://ai.google.dev/gemini-api/terms). Clonvo doesn’t store prompts or responses unless your organization turns on retention.

## Bring your own key

From Pay as you go up, a Google AI Studio key added under Provider keys is used for this model before your credits: $25,000 a month free, then 4% of list price. [How bring-your-own-key works](https://clonvo.com/docs/byok.md).
