# Cherry Studio

> Connect Cherry Studio to the LMU AI API and set up Claude, Chinese, and OpenAI model channels, with a direct overseas connection.

URL: https://docs.lmuai.ai/docs/tools/cherry











Cherry Studio is a multi-model AI desktop client.

***

## Download and install [#download-and-install]

* **Desktop official site**: [https://www.cherry-ai.com/](https://www.cherry-ai.com/)
* **Mobile (Android & iOS)**: [https://github.com/CherryHQ/cherry-studio-app/releases](https://github.com/CherryHQ/cherry-studio-app/releases)
  * Android: download the `.apk` file
  * iOS: download the `.ipa` file

The mobile install process is much like the desktop one — take a closer look and you'll find it.

***

## The two protocols [#the-two-protocols]

LMU AI supports two connection protocols; choose one based on the model type:

| Protocol            | Models                                               |
| ------------------- | ---------------------------------------------------- |
| **Anthropic**       | Claude models, Chinese models (DeepSeek, Qwen, etc.) |
| **OpenAI-Response** | OpenAI models (GPT-4, GPT-5, etc.)                   |

***

## Method 1: Anthropic protocol (Claude and Chinese models) [#method-1-anthropic-protocol-claude-and-chinese-models]

### Step 1 — Add a provider [#step-1--add-a-provider]

Go to Cherry Studio Settings → click Add Provider:

1. **Provider name**: anything you like (e.g. `LMU AI`)
2. **Provider type**: select **`Anthropic`** from the dropdown

<img alt="Add an Anthropic provider" src="__img0" />

### Step 2 — Enter the API details [#step-2--enter-the-api-details]

* **API Key**: enter the `sk-` key generated in the console
* **API Host**: enter the API relay request address

```
https://api.lmuai.ai
```

Click Fetch model list to retrieve the available models; the Claude series and Chinese models load automatically.

<img alt="Anthropic configuration complete" src="__img1" />

***

## Method 2: OpenAI-Response protocol (OpenAI models) [#method-2-openai-response-protocol-openai-models]

### Step 1 — Add a provider [#step-1--add-a-provider-1]

Go to Cherry Studio Settings → click Add Provider:

1. **Provider name**: anything you like (e.g. `LMU AI`)
2. **Provider type*&#x2A;: select &#x2A;*`OpenAI-Response`** from the dropdown (the second option, not the first OpenAI)

### Step 2 — Enter the API details [#step-2--enter-the-api-details-1]

* **API Key**: enter the `sk-` key generated in the console
* **API Host**: enter the API relay request address

```
https://api.lmuai.ai
```

Click Fetch model list to retrieve the available models; the GPT series loads automatically.

<img alt="OpenAI-Response configuration complete" src="__img2" />

Once configured, you can select and use the model in the chat screen:

<img alt="The chat screen" src="__img3" />

***

## Notes [#notes]

* The same API Key and API Host work for both protocols; just add two providers as needed
* For the OpenAI-Response type, be sure to pick the second option, not the first OpenAI
* Enter the API relay address as the API address; do not enter the official OpenAI or Anthropic address