The Studio Server connects to an Embedding Service to generate semantic embeddings for user queries. These embeddings are used to perform vector similarity searches against the indexed content embeddings in the search engine.

For details on the supported embedding models and their configuration, refer to Section 4.1.4.1, “Embedding Service Configuration”.

Studio Server Specific Configuration: All configuration properties specific to the Studio Server semantic search feature start with studio.rest.search-service.semantic.*. For a full list, see Section 3.4, “Studio Properties” in Deployment Manual.

Studio pulls the semantic search configuration from the Studio Server to enable semantic search for users. No additional configuration is required in Studio.