Quickly Building Your Own AI Assistant Using a ModelArts Studio (MaaS) DeepSeek API and Cherry Studio

Operation Scenarios

Large language models (LLMs) are now essential for advancing natural language processing applications. The DeepSeek series offers LLMs designed for efficiency and strong performance.

To simplify their use, several tools and platforms have been developed. Among them, Cherry Studio stands out as a versatile open-source desktop application compatible with Windows, macOS, and Linux. It combines popular LLMs like OpenAI, DeepSeek, and Gemini, enabling both cloud-based and local operations. Its features include dialog knowledge bases, AI painting, translation, and seamless model switching.

MaaS integrates DeepSeek models into its platform and enhances their accuracy and efficiency using AI Compute Service. Developers can test these models through API calls.

This guide explains how to access Cherry Studio with MaaS DeepSeek APIs to create your own AI assistant swiftly.

Solution Architecture

Figure 1 Solution architecture

• The user submits a request through Cherry Studio.
• Cherry Studio sends the request to MaaS DeepSeek.
• MaaS DeepSeek processes the request and returns the results to Cherry Studio.
• Cherry Studio optimizes the results and returns them to the user.

Billing

Model inference in MaaS uses compute and storage resources, which are billed. Compute resources are billed for running model services. Storage resources are billed for storing data in OBS. Using a built-in MaaS service charges you based on token count. For more information, see Model Inference Billing Items.

Prerequisites

• You have created a Huawei Cloud account. For details, see Signing Up for a HUAWEI ID and Enabling Huawei Cloud Services.
• You have completed ModelArts agency authorization. For details, see Configuring ModelArts Studio (MaaS) Access Authorization.

Step 1: Downloading and Installing Cherry Studio

Download Cherry Studio from the official website or open-source address.

Step 2: Obtaining MaaS DeepSeek Interconnection Information

Obtain key details about the target MaaS DeepSeek model service for connecting with Cherry Studio.

1. Create an API key for authentication when calling the MaaS DeepSeek model service.

   You can create up to 30 keys. Each key is displayed only once after creation. Keep it secure. If the key is lost, it cannot be retrieved. In this case, create a new API key.

   1. Log in to the ModelArts Studio (MaaS) console and select CN-Hong Kong from the top navigation bar.
   2. In the navigation pane, choose API Key Management.
   3. On the API Key Management page, click Create API Key, enter the tag and description, and click OK.

      The tag and description cannot be modified after the key is created.

      Table 1 Parameters

      Parameter	Description
      Tag	Tag of the API key. The tag must be unique. The tag can contain 1 to 100 characters. Only letters, digits, underscores (_), and hyphens (-) are allowed.
      Description	Description of the custom API key. The value can contain 1 to 100 characters.

   4. In the Your Key dialog box, copy the key and store it securely.
   5. After the key is saved, click Close.

      After you click Close, the key cannot be viewed again.

2. Obtain the API URL and model name of the MaaS DeepSeek model service.

   The following uses a deployed service as an example. You can also use commercial services to call APIs. For details, see Subscribing to a Built-in Commercial Service in ModelArts Studio (MaaS).

   1. In the navigation pane of ModelArts Studio (MaaS) console, choose Real-Time Inference.
   2. Click the My Services tab and click Deploy Model in the upper right corner to create a model service. For details, see Deploying a Model Service in ModelArts Studio (MaaS).
   3. Choose More > View Call Description in the Operation column of the target running model service.
   4. Check the basic API URL and model name on the page. You will need these for the next steps in configuring Cherry Studio.

Step 3: Configuring MaaS DeepSeek in Cherry Studio

1. Add a MaaS provider in Cherry Studio.
   1. In the lower left corner of the Cherry Studio client, click the settings icon and click Add in Model Provider.
      Figure 2 Adding a provider
      
   2. Set the provider name and provider type and click OK.

      Table 2 Parameters for adding a provider

      Parameter	Description
      Provider Name	Enter MaaS, which can be changed as needed.
      Provider Type	Set this parameter to OpenAI.

2. Add the MaaS DeepSeek API key and API URL in Cherry Studio.
   1. In the lower left corner of the Cherry Studio client, click the settings icon.
   2. Click MaaS and configure the API key and API URL.

      Parameter	Description
      API Key	API key created in Step 2: Obtaining MaaS DeepSeek Interconnection Information.
      API Host	Basic API URL of the MaaS service obtained in Step 2: Obtaining MaaS DeepSeek Interconnection Information. Remove /v1/chat/completions from the URL and enter it.

3. Add a model to Cherry Studio.
   1. In the Models area, click Add.
   2. Set the model ID, model name, and group name, and click Add Model.

      Parameter	Description
      Model ID	Model name obtained in Step 2: Obtaining MaaS DeepSeek Interconnection Information
      Model	Custom model name
      Group Name	Custom group name

Step 4: Using MaaS DeepSeek in Cherry Studio

1. In the navigation pane of Cherry Studio, click and select the configured model.
   Figure 3 Selecting a model
   
2. Enter message in the text box to start a conversation.

   Choose the model name in the top menu to switch models.

   Figure 4 Model Q&A example
   

FAQs

• How Long Does It Take for an API Key to Become Valid After It Is Created in MaaS?

  A MaaS API key becomes valid a few minutes after creation.

• How Do I Connect MaaS to Cherry Studio, Cursor, and Cline?

  MaaS integrates DeepSeek models and supports AI development on multiple platforms. For details, see Using ModelArts Studio (MaaS) DeepSeek API to Build AI Applications.