[DOCS]: Update semantic_text documentation (elastic#136482)

Author: charlotte-hoblik

docs/reference/query-languages/query-dsl/query-dsl-semantic-query.md

@@ -13,8 +13,18 @@ applies_to:
 We don't recommend this legacy query type for _new_ projects. Use the match query (with [QueryDSL](/reference/query-languages/query-dsl/query-dsl-match-query.md) or [ESQL](/reference/query-languages/esql/functions-operators/search-functions.md#esql-match)) instead. The semantic query remains available to support existing implementations.
 ::::
 
-The `semantic` query type enables you to perform [semantic search](docs-content://solutions/search/semantic-search.md) on data stored in a [`semantic_text`](/reference/elasticsearch/mapping-reference/semantic-text.md) field.
+The `semantic` query type enables you to perform [semantic search](docs-content://solutions/search/semantic-search.md) on data stored in a [`semantic_text`](/reference/elasticsearch/mapping-reference/semantic-text.md) field. This query accepts natural-language text and uses the field’s configured inference endpoint to generate a query embedding and match documents.
 
+For an overview of all query options available for `semantic_text` fields, see [Querying `semantic_text` fields](/reference/elasticsearch/mapping-reference/semantic-text.md#querying-semantic-text-fields).
+
+## Inference endpoint selection
+
+The target field of `semantic` query must be mapped as `semantic_text` and associated with an inference endpoint. At query time, the inference endpoint is chosen as follows:
+- If `search_inference_id` is defined, the semantic query uses that endpoint to embed the query.
+- If no `search_inference_id` is defined, `inference_id` is used for both indexing and search.
+- If no endpoint is specified at mapping, `inference_id` defaults to `.elser-2-elasticsearch`.
+
+The underlying vector type (dense or sparse) is determined by the endpoint automatically. No extra query parameters are required.
 
 ## Example request [semantic-query-example]
 

docs/reference/query-languages/query-dsl/query-dsl-sparse-vector-query.md

@@ -208,6 +208,54 @@ GET my-index/_search
 ```
 %  TEST[skip: Requires inference]
 
+## Example query on a `semantic_text` field
+
+You can also run a `sparse_vector` query directly on a [`semantic_text`](/reference/elasticsearch/mapping-reference/semantic-text.md) field. In this case Elasticsearch automatically uses the inference endpoint configured in the field mapping to expand the query into sparse tokens.
+
+First, create an index with a `semantic_text` field:
+
+```console
+PUT /my-semantic-sparse-index
+{
+  "mappings": {
+    "properties": {
+      "title": { "type": "text" },
+      "content_semantic": {
+        "type": "semantic_text",
+        "inference_id": ".elser-2-elasticsearch"
+      }
+    }
+  }
+}
+```
+
+Index some example documents:
+
+```console
+POST /my-semantic-sparse-index/_bulk
+{ "index": { "_index": "my-semantic-sparse-index", "_id": "1" } }
+{ "title": "Best surfing spots", "content_semantic": "Hawaii has world-class surfing with warm water and consistent swells." }
+{ "index": { "_index": "my-semantic-sparse-index", "_id": "2" } }
+{ "title": "City breaks", "content_semantic": "Paris offers museums, cafés, and beautiful architecture." }
+{ "index": { "_index": "my-semantic-sparse-index", "_id": "3" } }
+{ "title": "Learning to surf", "content_semantic": "Beginners often start on longboards at gentle beach breaks." }
+```
+
+Then query with `sparse_vector` against the `semantic_text` field:
+
+```console
+GET my-semantic-sparse-index/_search
+{
+  "size": 3,
+  "query": {
+    "sparse_vector": {
+      "field": "content_semantic",
+      "query": "best places to surf as a beginner" 
+    }
+  }
+}
+```
+
 
 ## Example ELSER query with pruning configuration and rescore [sparse-vector-query-with-pruning-config-and-rescore-example]