Platform For AI:Configure connections

Connection type

Key parameter description

General LLM Service

Configure a connection to an LLM service deployed on PAI-EAS or a third-party platform. For more information about deployment, see Model deployment and training.

• Service Provider:

  • PAI-EAS model service: Select this option to use a PAI-EAS model service. From the EAS Service list, select your deployed LLM service. The base_url and api_key are automatically set to the service's Virtual Private Cloud (VPC) endpoint and token, respectively.

  • Third-party model service: Select this option to use a third-party model service. For example, if you are using the official DeepSeek service, the base_url is https://api.deepseek.com from the DeepSeek official website.

• Model Name: When you deploy a model from Model Gallery as an EAS service, you can find the model name on the model details page. To access this page, click the model card on the Model Gallery page.

  

• Tool Call: Indicates whether the model supports the Tool Call feature of the OpenAI API. When using an Agent node, you must select a model that supports Tool Call.

General Embedding Model Service

Configure a connection to an Embedding model service deployed on PAI-EAS or a third-party platform. For more information, see Model deployment and training. The configuration process is the same as that for the General LLM Service.

General Multimodal Embedding Model Service

Configure a connection to a multimodal Embedding model service deployed on PAI-EAS or a third-party platform. For more information, see Model deployment and training. The configuration process is the same as that for the General LLM Service.

Alibaba Cloud Model Studio Service

Configure connections for model services on the Alibaba Cloud Model Studio platform. Key parameter description:

api_key: Access Alibaba Cloud Model Studio-API-KEY to create or obtain the API key.

DeepSeek Model Service

Configure a connection to a model service deployed on the Alibaba Cloud Model Studio platform.

api_key: Go to DeepSeek official website to create or find your API key.

AI Search Open Platform Model Service

Configure the model service connection for AI Search Open Platform (Note: You must complete the prerequisites Activate AI Search Open Platform first):

• endpoint: See Query service endpoint to obtain the service invocation address.

• api_key: See Manage API key to create or obtain the API key.

Multimodal VL Intelligent Annotation Model

Use this connection for intelligent labeling of multimodal data. It only supports model services deployed on PAI-EAS.

Connection type

Key parameter description

Search and Analysis Service - Elasticsearch

• url: the access address of the Elasticsearch cluster in the http://<private address>:<private port> format. You can obtain the private address and private port in the Basic Information section of the Elasticsearch cluster details page on the Elasticsearch console.

• username/password: The default username for Elasticsearch is elastic, and the password is the login password you set when you create the Elasticsearch cluster. If you forget the login password, you can reset the cluster access password.

Vector database-Milvus

• uri: the access address of the Milvus instance in the http://<Milvus internal access address> format. Example: http://c-b1c5222fba****-internal.milvus.aliyuncs.com. You can obtain the Milvus internal access address in the Access Address section of the instance details page on the Milvus console.

• token: the username and password for logging on to the Milvus instance in the <yourUsername>:<yourPassword> format.

• database: the database name. When you create a Milvus instance, the system automatically creates a database named default. You can also manually create a database.

Vector database-DashVector

• endpoint: Access the DashVector console, and obtain the VPC endpoint in the Access Method section on the cluster details page.

• api_key: To obtain the API key, see Manage API keys.

ApsaraDB RDS for MySQL

• host/port: To obtain the host (internal address) and port (internal port), see View and manage instance endpoints and ports.

• database: the database name set when you create a database.

• user/password: the account and password set when you create a database account.

Connection type

Key parameter description

Function AI

Use this to configure an MCP Service deployed through Alibaba Cloud Function AI.

Select an MCP service from the dropdown list to automatically populate the connection parameters. The url defaults to the public address of the MCP Service.

Custom

Use this to configure an MCP Service that you have deployed yourself or that is hosted on a third-party platform.

• transport: The transport protocol for communicating with the MCP Service. This service supports SSE and StreamableHTTP.

• url: The full endpoint URL for accessing the MCP Service.

• authentication: The authentication method for the MCP Service. You can use the following options:

  • BearerAuth: Use a bearer token for authentication. You must enter a valid access token in the token field.

  • Other: You must provide an auth_header_name to specify the request header that contains the token

  • None: This option requires no authentication.