This is the first article in the tutorial series on getting started with the MCP client: "Tutorial on Installing and Configuring Cherry Studio for MCP Client Beginners".

MCP (Model Context Protocol) is an open protocol launched by Anthropic in November 2024. It aims to achieve seamless integration between large language model (LLM) applications and external data sources, tools, and services. It provides a standardized method for AI models, enabling any large language model to easily connect to various data sources and tools for seamless access and processing of information, just like a USB-C interface for AI applications.

The core architecture of MCP is the "client-server" model:

• MCP Client: A connector embedded in AI applications, establishing a one-to-one connection with a specific MCP server, acting as a bridge between the model and the server. The client is responsible for discovering available services, sending call requests, and obtaining results.
• This article provides a detailed tutorial by AIbase on installing and configuring Cherry Studio, which is part of the tutorial series on getting started with the MCP client.

I. Introduction to Cherry Studio

Cherry Studio is a powerful MCP client that supports access to multiple large model APIs and is fully compatible with the MCP protocol. It allows AI models to call various tools to perform tasks. It features a simple interface and easy-to-use operation, supporting multiple operating systems such as Windows, macOS, and Linux. As a free tool for personal use, Cherry Studio has gained popularity among AI enthusiasts and professional users recently. In this tutorial, AIbase will teach you step by step how to install and configure Cherry Studio for getting started with the MCP client.

II. Client Installation

1. Download Cherry Studio

First, we need to visit the official website of Cherry Studio to download the client. The official website is: https://cherry-ai.com/.

2. Installation Process

After the download is completed, run the installation program. During the installation process, pay attention to the following options:

• Install for all users: You can choose the installation folder, and the default is the C drive folder. If you create a desktop shortcut and a start menu entry, all users of the computer can find them on the desktop and in the start menu, but user data is still stored in their respective user folders.
• Install only for me: It will be directly installed in the C drive user folder, and other users cannot find it on the desktop or in the start menu.

For ordinary users, the difference between the two options is not significant.

III. Environment Configuration

The latest version of Cherry Studio has integrated the tool installation function required for MCP. However, due to network issues, the automatic installation may not be smooth. The following introduces the manual installation method:

1. Install the UV and Bun Environments

Cherry Studio uses the built-in UV and Bun tools to run the MCP service. Make sure these tools are installed correctly:

Open Cherry Studio, click "Settings" in the lower left corner, select "MCP Server", and check if there is a prompt in the upper right corner of the page indicating that UV or Bun needs to be installed.

After the installation is completed, restart Cherry Studio, enter the MCP server settings, and the red triangle in the upper right corner should turn into a green checkmark, indicating that the tools have been installed successfully.

IV. Configure API Services

To use Cherry Studio to call large language models, you need to configure API services. Enter the settings interface: Click the small gear icon in the lower left corner of the interface to enter the settings interface.

1. Click the "Manage" button at the bottom, and you can see dozens of large models that can be called.

2. Select models with a "wrench" icon. These models support function calls and can use MCP services.

Currently available free or free-tier API platforms include:

• Domestic: Silicon Valley Flow, Volcengine, Alibaba Bailing, Tencent Yuanbao, Wuwen Cangqiong, etc.
• Overseas: Google Gemini, Grok, etc.

On the configuration interface of the model service, you can see that the page provides access to a wide range of large language models, covering almost all current mainstream platforms. You can choose according to your actual situation.

3. Create an API Key

To accommodate beginner users, this tutorial selects the model service provided by the "OpenRouter" service.

Then, go to the model service platform to apply for an API key. In this article, we take OpenRouter as an example, so in the example, enter OpenRouter and go to the console. (Official website: https://openrouter.ai/)

Click the "api key" option on the left side of the OpenRouter page.

Click the "New" button in the upper right corner, enter a description to facilitate later management of the key, and complete the creation of the key.

Click the transparently displayed key string, and the key will be automatically copied.

Paste the key: Return to the Studio interface, paste the copied key, and complete the key configuration.

V. Configure the MCP Server

There are three ways to configure the MCP server in Cherry Studio: search for MCP, edit MCP configuration, and more MCP.

Here, AIbase takes adding a remote MCP service (SSE type) as an example to demonstrate how to configure the MCP server

1. Search for MCP Services

Taking the addition of the web page crawling service fetch as an example: Visit the Magic Tower ModelScope MCP Square (https://modelscope.cn/mcp), search for, and select the fetch service.

Copy the SSE URL and return to Cherry Studio. The URL is as follows:

https://mcp-979bdfce-2a57-47a9.api-inference.modelscope.cn/sse

2. Add an MCP Server

Go to the "Settings" -> "MCP Server" page and click "Add Server".

Fill in the name (e.g., "fetch"), select the type as "SSE", paste the URL address, and click OK.

SSE-type services run on remote servers. They are easy to configure and do not require a local environment, but they cannot access local resources. They are suitable for scenarios such as obtaining cloud data and calling online APIs.

VI. Use MCP Services

After configuring the MCP services, you can start using them.

1. Open the Cherry Studio main interface, click to add an assistant or use the default assistant. Here, we directly use the default assistant.

2. Click the model service switch at the top and select a model that supports function calls (with a wrench icon).

3. Confirm that the MCP service switch at the bottom of the chat box is turned on. For example, if we have added the fetch MCP server above, the icon will turn green after selection.

4. Finally, ask the AI a question, and it will use the MCP service to complete the task.

Usage example:

I provide a URL for an AIbase daily report and ask the MCP service to extract all the sub - headings in the report. The result is as follows:

VII. Usage Tips and Precautions

When multiple MCP functions are similar: The instructions should be clear; otherwise, the AI may use the wrong tool.

Differences among different models: Different models have different levels of support for MCP. The domestic Qianwen model has better support.

Network issues: If there is a connection timeout or a closing error when adding a service, it may be a network issue. Try several times.

Permission issues: When using local file services, ensure that Cherry Studio has sufficient permissions to access the specified directory.

API quota: When using third - party APIs (such as Gaode Maps, Tavily Search), pay attention to the usage of the free quota.

Summary

As a powerful MCP client, Cherry Studio gives AI more powerful capabilities through the support of the MCP protocol. This article by AIbase details the installation and configuration of Cherry Studio and the usage of MCP service configuration, hoping to help you better utilize AI tools to improve work efficiency.

As the MCP ecosystem continues to improve, we can expect more feature - rich MCP services to emerge, making AI assistants more intelligent and practical. Through the combination of Cherry Studio and MCP, we can already experience the powerful potential of future AI assistants.

Official documentation: https://docs.cherry-ai.com/advanced-basic/mcp

MCP Client Getting Started Tutorial Series

《A Beginner's Guide to MCP for Novices: Learn to Use MCP Services from Scratch after Reading This Article》

《Tutorial on Installing and Configuring Cherry Studio for MCP Client Beginners》

《Tutorial on Installing and Configuring ChatWise for MCP Client Beginners》

《Tutorial on Installing and Configuring Cursor for MCP Client Beginners》

《Tutorial on Installing and Configuring Trae for MCP Client Beginners》

《Tutorial on Installing and Configuring Claude for MCP Client Beginners》