> ## Documentation Index
> Fetch the complete documentation index at: https://docs.viviai.cc/llms.txt
> Use this file to discover all available pages before exploring further.

# Cherry Studio

> Connect Cherry Studio to ViVi-API.

Have an API key from the [ViVi-API dashboard](https://api.viviai.cc/dashboard) ready, and confirm that its group supports your chosen model. Back up any existing configuration before merging the fields below.

**1. Open model settings**

Go to **Settings → Model providers** and find the option to add a provider or model. Labels may vary by version.

**2. Add a provider**

Enter a recognizable name, such as `ViVi-API`, and select the provider type for your protocol.

* General chat: Choose an OpenAI-compatible provider.
* Native Gemini or image generation: Choose Gemini and a suitable model.
* Native Claude: Choose Anthropic.

**3. Enter your key and URL, then add a model**

Enter your ViVi-API key and the appropriate API URL. OpenAI-compatible connections typically use `https://api.viviai.cc/v1`. For other protocols, follow the app's field instructions and avoid duplicate version paths.

Click **Manage / Add model**, enter the complete model ID, and enable it.

**4. Select the model in chat**

Return to the home or chat screen, select the model you added, and send a short test message.

## Nanobanana Pro: 2K / 4K settings

Some Cherry Studio versions do not fully support Nanobanana parameters. If your model supports high-resolution output, add this custom parameter in the model settings:

| Setting | Value |
| - | - |
| Parameter name | `imageConfig` |
| Type | `JSON` |

```json imageConfig theme={null}
{
  "aspectRatio": "9:16",
  "imageSize": "4K"
}
```

Adjust the aspect ratio as needed. For 2K output, change `imageSize` to `2K`. Whether the setting takes effect depends on the model, channel, and whether the client passes it correctly.

You can also use [Nanobanana online](https://nano.viviai.cc) and enter your key in the upper-right corner.

See the [Integration overview](https://docs.viviai.cc/en/scenarios.md) and [FAQ](https://docs.viviai.cc/en/faq.md) for more help.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.