> For the complete documentation index, see [llms.txt](https://docs.userouter.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.userouter.ai/studio/images.md).

# Images

Choose a model, size and quality, and see the price before you generate.

## Make an image

1. Open **Studio → Image**.
2. Pick a model and settings.
3. Enter a prompt (up to **4,000 characters**) and choose **1–4 images**.
4. Check the price shown, then generate.

Images usually arrive in seconds, each with a **Download** link.

## Models and settings

| Model                                            | Made by | Settings in Studio                                |
| ------------------------------------------------ | ------- | ------------------------------------------------- |
| GPT Image 2 (default)                            | OpenAI  | Aspect ratio; quality auto / medium / high        |
| GPT Image 2.5 Flare and Sunburst *(coming soon)* | OpenAI  | Aspect ratio; quality up to **xhigh** and **max** |
| Nano Banana 2                                    | Google  | Aspect ratio; resolution 512 / 1K / 2K / 4K       |

**Aspect ratios in Studio:**

| Aspect | Output size |
| ------ | ----------- |
| 1:1    | 1024 × 1024 |
| 3:2    | 1536 × 1024 |
| 2:3    | 1024 × 1536 |
| 16:9   | 2048 × 1152 |
| 9:16   | 1152 × 2048 |

The default quality is **medium**. **Auto** is priced the same as medium.

## What an image costs

You pay per image. The price depends on the model, quality and size (GPT Image) or resolution (Nano Banana 2), and is multiplied by the number of images.

Example: 4 images on GPT Image 2, medium, 1:1 = 4 × 51 = **204 credits**.

If only some images are delivered (for example 3 of 4), you pay for the ones you got and the rest is refunded. See [Refunds](/credits-and-billing/refunds.md).

Full prices: [Model catalogue](/get-started/models.md#image).

## Link to Studio with a prompt

You can open Studio with a model and prompt filled in:

```
https://userouter.ai/studio?model=gpt-image-2&prompt=Neon%20koi%20in%20a%20rainy%20alley
```

`model` takes an image model ID and `prompt` is trimmed to 4,000 characters.

Generated images are stored at public links. See [Privacy and your data](/help/privacy.md).