AI Search Open Platform offers component-based model services for intelligent search and retrieval-augmented generation (RAG) scenarios. These services include document parsing, document segmentation, text embedding, retrieval, sorting, and Large Language Model (LLM) responses. This topic uses the text embedding service as an example to demonstrate how to use the service in the AI Search Open Platform console or by calling an API or SDK.
Prerequisites
-
Activate the AI Search Open Platform service. For more information, see Activate the service.
-
If you call the service using an API or SDK, obtain an API key for authentication. For more information, see Manage API keys.
Try the service in the console
-
Log on to AI Search Open Platform.
-
Select the Shanghai region and switch to AI Search Open Platform.
Note-
Currently, AI Search Open Platform is available only in the China (Shanghai) and Germany (Frankfurt) regions.
-
Users in the China (Hangzhou), China (Shenzhen), China (Beijing), China (Zhangjiakou), and China (Qingdao) regions can invoke AI Search Open Platform services across regions using VPC endpoints.
-
-
In the navigation pane on the left, choose Experience Center.
-
Set Service Category to Text Embedding and select a specific embedding service type.
-
The document type is for embedding knowledge base documents, and the query type is for embedding queries.
-
Click Get Result. The system then calls the service to embed the text.
-
Result: Displays the embedding result.
-
Result Source Code: View the result response code. You can download the code to your local machine by clicking Copy Code or Download File.
-
Sample Code: Provides sample code for curl and Python calls.
-
Try the service using an API or SDK
For more information, see Text embedding and Sparse vector.