From 8b03ce93a833b0f2c2a10dab311ea862671e962d Mon Sep 17 00:00:00 2001
From: Jim Ferenczi <jim.ferenczi@elastic.co>
Date: Fri, 25 Jul 2025 10:10:14 +0200
Subject: [PATCH 01/16] Enable `exclude_source_vectors` by default for new
 indices

This commit sets `index.mapping.exclude_source_vectors` to `true` by default
for newly created indices. When enabled, vector fields (`dense_vector`,
`sparse_vector`, `rank_vector`) are excluded from `_source` on disk and are
not returned in API responses unless explicitly requested.

The change improves indexing performance, reduces storage size, and avoids
unnecessary payload bloat in responses. Vector values continue to be rehydrated
transparently for partial updates, reindex, and recovery.

Existing indices are not affected and continue to store vectors in `_source`
by default.
---
 .../mapping-reference/dense-vector.md         | 71 ++++++++++++++++
 .../mapping-reference/rank-vectors.md         | 70 ++++++++++++++-
 .../mapping-reference/sparse-vector.md        | 76 +++++++++++++++--
 .../reindex/ReindexBasicTests.java            |  7 +-
 .../reindex/UpdateByQueryBasicTests.java      |  4 +-
 .../test/get/100_synthetic_source.yml         | 14 +++
 .../240_source_synthetic_dense_vectors.yml    |  4 +-
 .../250_source_synthetic_sparse_vectors.yml   |  4 +-
 .../test/search.vectors/90_sparse_vector.yml  | 52 ++++++++----
 .../common/settings/IndexScopedSettings.java  |  6 +-
 .../elasticsearch/index/IndexSettings.java    |  7 +-
 .../elasticsearch/index/IndexVersions.java    |  1 +
 .../index/get/ShardGetService.java            |  4 +-
 .../index/mapper/DocumentParser.java          |  2 +-
 .../vectors/DenseVectorFieldMapper.java       | 24 +++---
 .../vectors/SparseVectorFieldMapper.java      | 85 ++++---------------
 .../action/search/SearchCapabilities.java     |  8 +-
 .../vectors/DenseVectorFieldMapperTests.java  | 17 +---
 .../vectors/SparseVectorFieldMapperTests.java | 15 +---
 .../SyntheticVectorFieldsRecoveryTests.java   |  4 +-
 .../SyntheticVectorsMapperTestCase.java       |  9 +-
 .../index/shard/ShardGetServiceTests.java     | 15 +++-
 .../xpack/rank/vectors/RankVectorsPlugin.java |  4 +-
 .../mapper/RankVectorsFieldMapper.java        | 20 ++---
 .../test/ml/sparse_vector_search.yml          | 29 ++++---
 .../rank_vectors_synthetic_vectors.yml        |  4 +-
 26 files changed, 353 insertions(+), 203 deletions(-)

diff --git a/docs/reference/elasticsearch/mapping-reference/dense-vector.md b/docs/reference/elasticsearch/mapping-reference/dense-vector.md
index 985f62f37eec9..45dc923859c6b 100644
--- a/docs/reference/elasticsearch/mapping-reference/dense-vector.md
+++ b/docs/reference/elasticsearch/mapping-reference/dense-vector.md
@@ -95,6 +95,77 @@ PUT my-index-2
 
 {{es}} uses the [HNSW algorithm](https://arxiv.org/abs/1603.09320) to support efficient kNN search. Like most kNN algorithms, HNSW is an approximate method that sacrifices result accuracy for improved speed.
 
+## Accessing `dense_vector` fields in search responses
+
+By default, `dense_vector` fields are **not included in `_source`** in responses from the `_search`, `_msearch`, `_get`, and `_mget` APIs.
+This helps reduce response size and improve performance, especially in scenarios where vectors are used solely for similarity scoring and not required in the output.
+
+To retrieve vector values explicitly, you can use:
+
+* **The `fields` option** to request specific vector fields directly:
+
+```console
+POST my-index-2/_search
+{
+  "fields": ["my_vector"]
+}
+```
+
+- **The `_source.exclude_vectors` flag** to re-enable vector inclusion in `_source` responses:
+
+```console
+POST my-index-2/_search
+{
+  "_source": {
+    "exclude_vectors": false
+  }
+}
+```
+
+### Storage behavior and `_source`
+
+By default, `dense_vector` fields are **not stored in `_source`** on disk. This is also controlled by the index setting `index.mapping.exclude_source_vectors`.
+This setting is enabled by default for newly created indices and can only be set at index creation time.
+
+When enabled:
+
+* `dense_vector` fields are removed from `_source` and the rest of the `_source` is stored as usual.
+* If a request includes `_source` and vector values are needed (e.g., during recovery or reindex), the vectors are **rehydrated** from their internal format.
+
+This setting is **compatible with synthetic `_source`**, where the entire `_source` document is reconstructed from columnar storage. In full synthetic mode, no `_source` is stored on disk, and all fields — including vectors — are rebuilt when needed.
+
+### Rehydration and precision
+
+When vector values are rehydrated (e.g., for reindex, recovery, or explicit `_source` requests), they are restored from their internal format. Internally, vectors are stored at **float precision**, so if they were originally indexed as higher-precision types (e.g., `double` or `long`), the rehydrated values will have **reduced precision**. This lossy representation is intended to save space while preserving search quality.
+
+### Storing original vectors in `_source`
+
+If you want to preserve the **original vector values exactly as they were provided**, you can re-enable vector storage in `_source`:
+
+```console
+PUT my-index-include-vectors
+{
+  "settings": {
+    "index.mapping.exclude_source_vectors": false
+  },
+  "mappings": {
+    "properties": {
+      "my_vector": {
+        "type": "dense_vector"
+      }
+    }
+  }
+}
+```
+
+When this setting is disabled:
+
+* `dense_vector` fields are stored as part of the `_source`, exactly as indexed.
+* The index will store both the original `_source` value and the internal representation used for vector search, resulting in **increased storage usage**.
+* **Vectors are once again returned in `_source` by default** in all relevant APIs, with no need to use `exclude_vectors` or `fields`.
+
+This configuration is appropriate when full source fidelity is required, such as for auditing or round-tripping exact input values.
+
 ## Automatically quantize vectors for kNN search [dense-vector-quantization]
 
 The `dense_vector` type supports quantization to reduce the memory footprint required when [searching](docs-content://solutions/search/vector/knn.md#approximate-knn) `float` vectors. The three following quantization strategies are supported:
diff --git a/docs/reference/elasticsearch/mapping-reference/rank-vectors.md b/docs/reference/elasticsearch/mapping-reference/rank-vectors.md
index 20683e89b164c..7490cee0760bf 100644
--- a/docs/reference/elasticsearch/mapping-reference/rank-vectors.md
+++ b/docs/reference/elasticsearch/mapping-reference/rank-vectors.md
@@ -108,11 +108,77 @@ $$$rank-vectors-element-type$$$
 `dims`
 :   (Optional, integer) Number of vector dimensions. Can’t exceed `4096`. If `dims` is not specified, it will be set to the length of the first vector added to the field.
 
+## Accessing `dense_vector` fields in search responses
 
-## Synthetic `_source` [rank-vectors-synthetic-source]
+By default, `dense_vector` fields are **not included in `_source`** in responses from the `_search`, `_msearch`, `_get`, and `_mget` APIs.
+This helps reduce response size and improve performance, especially in scenarios where vectors are used solely for similarity scoring and not required in the output.
 
-`rank_vectors` fields support [synthetic `_source`](mapping-source-field.md#synthetic-source) .
+To retrieve vector values explicitly, you can use:
 
+* **The `fields` option** to request specific vector fields directly:
+
+```console
+POST my-index-2/_search
+{
+  "fields": ["my_vector"]
+}
+```
+
+- **The `_source.exclude_vectors` flag** to re-enable vector inclusion in `_source` responses:
+
+```console
+POST my-index-2/_search
+{
+  "_source": {
+    "exclude_vectors": false
+  }
+}
+```
+
+### Storage behavior and `_source`
+
+By default, `rank_vectors` fields are **not stored in `_source`** on disk. This is also controlled by the index setting `index.mapping.exclude_source_vectors`.
+This setting is enabled by default for newly created indices and can only be set at index creation time.
+
+When enabled:
+
+* `rank_vectors` fields are removed from `_source` and the rest of the `_source` is stored as usual.
+* If a request includes `_source` and vector values are needed (e.g., during recovery or reindex), the vectors are **rehydrated** from their internal format.
+
+This setting is **compatible with synthetic `_source`**, where the entire `_source` document is reconstructed from columnar storage. In full synthetic mode, no `_source` is stored on disk, and all fields — including vectors — are rebuilt when needed.
+
+### Rehydration and precision
+
+When vector values are rehydrated (e.g., for reindex, recovery, or explicit `_source` requests), they are restored from their internal format. Internally, vectors are stored at **float precision**, so if they were originally indexed as higher-precision types (e.g., `double` or `long`), the rehydrated values will have **reduced precision**. This lossy representation is intended to save space while preserving search quality.
+
+### Storing original vectors in `_source`
+
+If you want to preserve the **original vector values exactly as they were provided**, you can re-enable vector storage in `_source`:
+
+```console
+PUT my-index-include-vectors
+{
+  "settings": {
+    "index.mapping.exclude_source_vectors": false
+  },
+  "mappings": {
+    "properties": {
+      "my_vector": {
+        "type": "rank_vectors",
+        "dims": 128
+      }
+    }
+  }
+}
+```
+
+When this setting is disabled:
+
+* `rank_vectors` fields are stored as part of the `_source`, exactly as indexed.
+* The index will store both the original `_source` value and the internal representation used for vector search, resulting in **increased storage usage**.
+* **Vectors are once again returned in `_source` by default** in all relevant APIs, with no need to use `exclude_vectors` or `fields`.
+
+This configuration is appropriate when full source fidelity is required, such as for auditing or round-tripping exact input values.
 
 ## Scoring with rank vectors [rank-vectors-scoring]
 
diff --git a/docs/reference/elasticsearch/mapping-reference/sparse-vector.md b/docs/reference/elasticsearch/mapping-reference/sparse-vector.md
index 3a65ea4fc5ff8..595f9a1df0992 100644
--- a/docs/reference/elasticsearch/mapping-reference/sparse-vector.md
+++ b/docs/reference/elasticsearch/mapping-reference/sparse-vector.md
@@ -57,12 +57,6 @@ See [semantic search with ELSER](docs-content://solutions/search/semantic-search
 
 The following parameters are accepted by `sparse_vector` fields:
 
-[store](/reference/elasticsearch/mapping-reference/mapping-store.md)
-:   Indicates whether the field value should be stored and retrievable independently of the [_source](/reference/elasticsearch/mapping-reference/mapping-source-field.md) field. Accepted values: true or false (default). The field’s data is stored using term vectors, a disk-efficient structure compared to the original JSON input. The input map can be retrieved during a search request via the [`fields` parameter](/reference/elasticsearch/rest-apis/retrieve-selected-fields.md#search-fields-param). To benefit from reduced disk usage, you must either:
-
-    * Exclude the field from [_source](/reference/elasticsearch/rest-apis/retrieve-selected-fields.md#source-filtering).
-    * Use [synthetic `_source`](/reference/elasticsearch/mapping-reference/mapping-source-field.md#synthetic-source).
-
 index_options {applies_to}`stack: ga 9.1`
 :   (Optional, object) You can set index options for your  `sparse_vector` field to determine if you should prune tokens, and the parameter configurations for the token pruning. If pruning options are not set in your [`sparse_vector` query](/reference/query-languages/query-dsl/query-dsl-sparse-vector-query.md), Elasticsearch will use the default options configured for the field, if any.
 
@@ -96,6 +90,76 @@ This ensures that:
 * The tokens that are kept are frequent enough and have significant scoring.
 * Very infrequent tokens that may not have as high of a score are removed.
 
+## Accessing `dense_vector` fields in search responses
+
+By default, `dense_vector` fields are **not included in `_source`** in responses from the `_search`, `_msearch`, `_get`, and `_mget` APIs.
+This helps reduce response size and improve performance, especially in scenarios where vectors are used solely for similarity scoring and not required in the output.
+
+To retrieve vector values explicitly, you can use:
+
+* **The `fields` option** to request specific vector fields directly:
+
+```console
+POST my-index-2/_search
+{
+  "fields": ["my_vector"]
+}
+```
+
+- **The `_source.exclude_vectors` flag** to re-enable vector inclusion in `_source` responses:
+
+```console
+POST my-index-2/_search
+{
+  "_source": {
+    "exclude_vectors": false
+  }
+}
+```
+
+### Storage behavior and `_source`
+
+By default, `sparse_vector` fields are **not stored in `_source`** on disk. This is also controlled by the index setting `index.mapping.exclude_source_vectors`.
+This setting is enabled by default for newly created indices and can only be set at index creation time.
+
+When enabled:
+
+* `sparse_vector` fields are removed from `_source` and the rest of the `_source` is stored as usual.
+* If a request includes `_source` and vector values are needed (e.g., during recovery or reindex), the vectors are **rehydrated** from their internal format.
+
+This setting is **compatible with synthetic `_source`**, where the entire `_source` document is reconstructed from columnar storage. In full synthetic mode, no `_source` is stored on disk, and all fields — including vectors — are rebuilt when needed.
+
+### Rehydration and precision
+
+When vector values are rehydrated (e.g., for reindex, recovery, or explicit `_source` requests), they are restored from their internal format. Internally, vectors are stored at **float precision**, so if they were originally indexed as higher-precision types (e.g., `double` or `long`), the rehydrated values will have **reduced precision**. This lossy representation is intended to save space while preserving search quality.
+
+### Storing original vectors in `_source`
+
+If you want to preserve the **original vector values exactly as they were provided**, you can re-enable vector storage in `_source`:
+
+```console
+PUT my-index-include-vectors
+{
+  "settings": {
+    "index.mapping.exclude_source_vectors": false
+  },
+  "mappings": {
+    "properties": {
+      "my_vector": {
+        "type": "sparse_vector"
+      }
+    }
+  }
+}
+```
+
+When this setting is disabled:
+
+* `sparse_vector` fields are stored as part of the `_source`, exactly as indexed.
+* The index will store both the original `_source` value and the internal representation used for vector search, resulting in **increased storage usage**.
+* **Vectors are once again returned in `_source` by default** in all relevant APIs, with no need to use `exclude_vectors` or `fields`.
+
+This configuration is appropriate when full source fidelity is required, such as for auditing or round-tripping exact input values.
 
 ## Multi-value sparse vectors [index-multi-value-sparse-vectors]
 
diff --git a/modules/reindex/src/test/java/org/elasticsearch/reindex/ReindexBasicTests.java b/modules/reindex/src/test/java/org/elasticsearch/reindex/ReindexBasicTests.java
index 96c7ef49f6956..e8f8efe8114d5 100644
--- a/modules/reindex/src/test/java/org/elasticsearch/reindex/ReindexBasicTests.java
+++ b/modules/reindex/src/test/java/org/elasticsearch/reindex/ReindexBasicTests.java
@@ -23,7 +23,6 @@
 import java.util.Map;
 import java.util.stream.Collectors;
 
-import static org.elasticsearch.index.IndexSettings.SYNTHETIC_VECTORS;
 import static org.elasticsearch.index.query.QueryBuilders.termQuery;
 import static org.elasticsearch.test.hamcrest.ElasticsearchAssertions.assertAcked;
 import static org.elasticsearch.test.hamcrest.ElasticsearchAssertions.assertHitCount;
@@ -182,14 +181,13 @@ public void testReindexFromComplexDateMathIndexName() throws Exception {
     }
 
     public void testReindexIncludeVectors() throws Exception {
-        assumeTrue("This test requires synthetic vectors to be enabled", SYNTHETIC_VECTORS);
         var resp1 = prepareCreate("test").setSettings(
-            Settings.builder().put(IndexSettings.INDEX_MAPPING_SOURCE_SYNTHETIC_VECTORS_SETTING.getKey(), true).build()
+            Settings.builder().put(IndexSettings.INDEX_MAPPING_SOURCE_EXCLUDE_SOURCE_VECTORS_SETTING.getKey(), true).build()
         ).setMapping("foo", "type=dense_vector,similarity=l2_norm", "bar", "type=sparse_vector").get();
         assertAcked(resp1);
 
         var resp2 = prepareCreate("test_reindex").setSettings(
-            Settings.builder().put(IndexSettings.INDEX_MAPPING_SOURCE_SYNTHETIC_VECTORS_SETTING.getKey(), true).build()
+            Settings.builder().put(IndexSettings.INDEX_MAPPING_SOURCE_EXCLUDE_SOURCE_VECTORS_SETTING.getKey(), true).build()
         ).setMapping("foo", "type=dense_vector,similarity=l2_norm", "bar", "type=sparse_vector").get();
         assertAcked(resp2);
 
@@ -237,5 +235,4 @@ public void testReindexIncludeVectors() throws Exception {
             searchResponse.decRef();
         }
     }
-
 }
diff --git a/modules/reindex/src/test/java/org/elasticsearch/reindex/UpdateByQueryBasicTests.java b/modules/reindex/src/test/java/org/elasticsearch/reindex/UpdateByQueryBasicTests.java
index 33c80e9138d28..2075b63862af6 100644
--- a/modules/reindex/src/test/java/org/elasticsearch/reindex/UpdateByQueryBasicTests.java
+++ b/modules/reindex/src/test/java/org/elasticsearch/reindex/UpdateByQueryBasicTests.java
@@ -24,7 +24,6 @@
 import java.util.Map;
 import java.util.stream.Collectors;
 
-import static org.elasticsearch.index.IndexSettings.SYNTHETIC_VECTORS;
 import static org.elasticsearch.index.query.QueryBuilders.termQuery;
 import static org.elasticsearch.test.hamcrest.ElasticsearchAssertions.assertAcked;
 import static org.elasticsearch.test.hamcrest.ElasticsearchAssertions.assertHitCount;
@@ -158,9 +157,8 @@ public void testMissingSources() {
     }
 
     public void testUpdateByQueryIncludeVectors() throws Exception {
-        assumeTrue("This test requires synthetic vectors to be enabled", SYNTHETIC_VECTORS);
         var resp1 = prepareCreate("test").setSettings(
-            Settings.builder().put(IndexSettings.INDEX_MAPPING_SOURCE_SYNTHETIC_VECTORS_SETTING.getKey(), true).build()
+            Settings.builder().put(IndexSettings.INDEX_MAPPING_SOURCE_EXCLUDE_SOURCE_VECTORS_SETTING.getKey(), true).build()
         ).setMapping("foo", "type=dense_vector,similarity=l2_norm", "bar", "type=sparse_vector").get();
         assertAcked(resp1);
 
diff --git a/rest-api-spec/src/yamlRestTest/resources/rest-api-spec/test/get/100_synthetic_source.yml b/rest-api-spec/src/yamlRestTest/resources/rest-api-spec/test/get/100_synthetic_source.yml
index a0061272a2c23..91a1d1bf9ef40 100644
--- a/rest-api-spec/src/yamlRestTest/resources/rest-api-spec/test/get/100_synthetic_source.yml
+++ b/rest-api-spec/src/yamlRestTest/resources/rest-api-spec/test/get/100_synthetic_source.yml
@@ -427,6 +427,11 @@ indexed dense vectors:
   - requires:
       cluster_features: ["gte_v8.5.0"]
       reason: introduced in 8.5.0
+      test_runner_features: [ capabilities ]
+      capabilities:
+        - method: GET
+          path: /_search
+          capabilities: [ exclude_source_vectors_setting ]
 
   - do:
       indices.create:
@@ -457,6 +462,8 @@ indexed dense vectors:
       get:
         index: test
         id:    1
+        _source_exclude_vectors: false
+
   - match: {_index: "test"}
   - match: {_id: "1"}
   - match: {_version: 1}
@@ -472,6 +479,11 @@ non-indexed dense vectors:
   - requires:
       cluster_features: ["gte_v8.5.0"]
       reason: introduced in 8.5.0
+      test_runner_features: [ capabilities ]
+      capabilities:
+        - method: GET
+          path: /_search
+          capabilities: [ exclude_source_vectors_setting ]
 
   - do:
       indices.create:
@@ -501,6 +513,8 @@ non-indexed dense vectors:
       get:
         index: test
         id:    1
+        _source_exclude_vectors: false
+
   - match: {_index: "test"}
   - match: {_id: "1"}
   - match: {_version: 1}
diff --git a/rest-api-spec/src/yamlRestTest/resources/rest-api-spec/test/search.vectors/240_source_synthetic_dense_vectors.yml b/rest-api-spec/src/yamlRestTest/resources/rest-api-spec/test/search.vectors/240_source_synthetic_dense_vectors.yml
index 414f6cfdad645..68f8c868b4e7e 100644
--- a/rest-api-spec/src/yamlRestTest/resources/rest-api-spec/test/search.vectors/240_source_synthetic_dense_vectors.yml
+++ b/rest-api-spec/src/yamlRestTest/resources/rest-api-spec/test/search.vectors/240_source_synthetic_dense_vectors.yml
@@ -5,7 +5,7 @@ setup:
       capabilities:
         - method: GET
           path: /_search
-          capabilities: [ synthetic_vectors_setting ]
+          capabilities: [ exclude_source_vectors_setting ]
   - skip:
       features: "headers"
 
@@ -13,8 +13,6 @@ setup:
       indices.create:
         index: test
         body:
-          settings:
-            index.mapping.synthetic_vectors: true
           mappings:
             properties:
               name:
diff --git a/rest-api-spec/src/yamlRestTest/resources/rest-api-spec/test/search.vectors/250_source_synthetic_sparse_vectors.yml b/rest-api-spec/src/yamlRestTest/resources/rest-api-spec/test/search.vectors/250_source_synthetic_sparse_vectors.yml
index 53f0cd33da7d3..8397b48866204 100644
--- a/rest-api-spec/src/yamlRestTest/resources/rest-api-spec/test/search.vectors/250_source_synthetic_sparse_vectors.yml
+++ b/rest-api-spec/src/yamlRestTest/resources/rest-api-spec/test/search.vectors/250_source_synthetic_sparse_vectors.yml
@@ -5,7 +5,7 @@ setup:
       capabilities:
         - method: GET
           path: /_search
-          capabilities: [ synthetic_vectors_setting ]
+          capabilities: [ exclude_source_vectors_setting ]
   - skip:
       features: "headers"
 
@@ -13,8 +13,6 @@ setup:
       indices.create:
         index: test
         body:
-          settings:
-            index.mapping.synthetic_vectors: true
           mappings:
             properties:
               name:
diff --git a/rest-api-spec/src/yamlRestTest/resources/rest-api-spec/test/search.vectors/90_sparse_vector.yml b/rest-api-spec/src/yamlRestTest/resources/rest-api-spec/test/search.vectors/90_sparse_vector.yml
index 0b65a69bf500e..b521b2866f9c5 100644
--- a/rest-api-spec/src/yamlRestTest/resources/rest-api-spec/test/search.vectors/90_sparse_vector.yml
+++ b/rest-api-spec/src/yamlRestTest/resources/rest-api-spec/test/search.vectors/90_sparse_vector.yml
@@ -387,8 +387,13 @@
 "sparse_vector synthetic source":
 
   - requires:
-      cluster_features: [ "mapper.source.mode_from_index_setting" ]
+      cluster_features: [ "mapper.source.mode_from_index_setting"]
       reason: "Source mode configured through index setting"
+      test_runner_features: [ capabilities, "close_to" ]
+      capabilities:
+        - method: GET
+          path: /_search
+          capabilities: [ exclude_vectors_param, exclude_source_vectors_setting ]
 
   - do:
       indices.create:
@@ -402,6 +407,18 @@
               ml.tokens:
                 type: sparse_vector
 
+  - do:
+      indices.create:
+        index: test_include_vectors
+        body:
+          settings:
+            index:
+              mapping.exclude_source_vectors: false
+          mappings:
+            properties:
+              ml.tokens:
+                type: sparse_vector
+
   - match: { acknowledged: true }
 
   - do:
@@ -421,7 +438,7 @@
 
   - do:
       index:
-        index: test
+        index: test_include_vectors
         id: "2"
         body:
           ml:
@@ -431,7 +448,7 @@
 
   - do:
       index:
-        index: test
+        index: test_include_vectors
         id: "3"
         body:
           ml:
@@ -446,20 +463,17 @@
       get:
         index: test
         id: "1"
+        _source_exclude_vectors: false
 
-  - match:
-      _source:
-        ml:
-          tokens:
-            running: 2.4097164
-            good: 2.170997
-            run: 2.052153
-            race: 1.4575411
-            for: 1.1908325
+  - close_to: { _source.ml.tokens.running: { value: 2.4097164, error: 0.01 } }
+  - close_to: { _source.ml.tokens.good: { value: 2.170997, error: 0.01 } }
+  - close_to: { _source.ml.tokens.run: { value: 2.052153, error: 0.01 } }
+  - close_to: { _source.ml.tokens.race: { value: 1.4575411, error: 0.01 } }
+  - close_to: { _source.ml.tokens.for: { value: 1.1908325, error: 0.01 } }
 
   - do:
       get:
-        index: test
+        index: test_include_vectors
         id: "2"
 
   - match:
@@ -467,7 +481,7 @@
 
   - do:
       get:
-        index: test
+        index: test_include_vectors
         id: "3"
 
   - match:
@@ -527,8 +541,14 @@
 "stored sparse_vector synthetic source":
 
   - requires:
-      cluster_features: [ "mapper.source.mode_from_index_setting", "mapper.sparse_vector.store_support" ]
       reason: "sparse_vector supports store parameter"
+      cluster_features: [ "mapper.source.mode_from_index_setting", "mapper.sparse_vector.store_support" ]
+      test_runner_features: [ capabilities, "close_to" ]
+      capabilities:
+        - method: GET
+          path: /_search
+          capabilities: [ exclude_vectors_param ]
+
 
   - do:
       indices.create:
@@ -567,6 +587,8 @@
       search:
         index: test
         body:
+          _source:
+            exclude_vectors: false
           fields: [ "ml.tokens" ]
 
   - match:
diff --git a/server/src/main/java/org/elasticsearch/common/settings/IndexScopedSettings.java b/server/src/main/java/org/elasticsearch/common/settings/IndexScopedSettings.java
index 9f4c5b80ccf23..db4278bf991d5 100644
--- a/server/src/main/java/org/elasticsearch/common/settings/IndexScopedSettings.java
+++ b/server/src/main/java/org/elasticsearch/common/settings/IndexScopedSettings.java
@@ -49,8 +49,6 @@
 import java.util.Map;
 import java.util.Set;
 
-import static org.elasticsearch.index.IndexSettings.SYNTHETIC_VECTORS;
-
 /**
  * Encapsulates all valid index level settings.
  * @see Property#IndexScope
@@ -243,9 +241,7 @@ public final class IndexScopedSettings extends AbstractScopedSettings {
         if (IndexSettings.DOC_VALUES_SKIPPER) {
             settings.add(IndexSettings.USE_DOC_VALUES_SKIPPER);
         }
-        if (SYNTHETIC_VECTORS) {
-            settings.add(IndexSettings.INDEX_MAPPING_SOURCE_SYNTHETIC_VECTORS_SETTING);
-        }
+        settings.add(IndexSettings.INDEX_MAPPING_SOURCE_EXCLUDE_SOURCE_VECTORS_SETTING);
         BUILT_IN_INDEX_SETTINGS = Collections.unmodifiableSet(settings);
     };
 
diff --git a/server/src/main/java/org/elasticsearch/index/IndexSettings.java b/server/src/main/java/org/elasticsearch/index/IndexSettings.java
index cd78d4323f44b..3912e6297e293 100644
--- a/server/src/main/java/org/elasticsearch/index/IndexSettings.java
+++ b/server/src/main/java/org/elasticsearch/index/IndexSettings.java
@@ -848,10 +848,9 @@ private static String getIgnoreAboveDefaultValue(final Settings settings) {
         Property.Final
     );
 
-    public static final boolean SYNTHETIC_VECTORS = new FeatureFlag("mapping_synthetic_vectors").isEnabled();
-    public static final Setting<Boolean> INDEX_MAPPING_SOURCE_SYNTHETIC_VECTORS_SETTING = Setting.boolSetting(
-        "index.mapping.synthetic_vectors",
-        false,
+    public static final Setting<Boolean> INDEX_MAPPING_SOURCE_EXCLUDE_SOURCE_VECTORS_SETTING = Setting.boolSetting(
+        "index.mapping.exclude_source_vectors",
+        settings -> String.valueOf(SETTING_INDEX_VERSION_CREATED.get(settings).onOrAfter(IndexVersions.EXCLUDE_SOURCE_VECTORS_DEFAULT)),
         Property.IndexScope,
         Property.Final
     );
diff --git a/server/src/main/java/org/elasticsearch/index/IndexVersions.java b/server/src/main/java/org/elasticsearch/index/IndexVersions.java
index 2e464afa72b76..dc102addb12d2 100644
--- a/server/src/main/java/org/elasticsearch/index/IndexVersions.java
+++ b/server/src/main/java/org/elasticsearch/index/IndexVersions.java
@@ -180,6 +180,7 @@ private static Version parseUnchecked(String version) {
     public static final IndexVersion SPARSE_VECTOR_PRUNING_INDEX_OPTIONS_SUPPORT = def(9_031_0_00, Version.LUCENE_10_2_2);
     public static final IndexVersion DEFAULT_DENSE_VECTOR_TO_BBQ_HNSW = def(9_032_0_00, Version.LUCENE_10_2_2);
     public static final IndexVersion MATCH_ONLY_TEXT_STORED_AS_BYTES = def(9_033_0_00, Version.LUCENE_10_2_2);
+    public static final IndexVersion EXCLUDE_SOURCE_VECTORS_DEFAULT = def(9_034_0_00, Version.LUCENE_10_2_2);
 
     /*
      * STOP! READ THIS FIRST! No, really,
diff --git a/server/src/main/java/org/elasticsearch/index/get/ShardGetService.java b/server/src/main/java/org/elasticsearch/index/get/ShardGetService.java
index 1e553e913379c..ac6e6b0fea718 100644
--- a/server/src/main/java/org/elasticsearch/index/get/ShardGetService.java
+++ b/server/src/main/java/org/elasticsearch/index/get/ShardGetService.java
@@ -60,7 +60,7 @@
 import java.util.function.Function;
 import java.util.stream.Collectors;
 
-import static org.elasticsearch.index.IndexSettings.INDEX_MAPPING_SOURCE_SYNTHETIC_VECTORS_SETTING;
+import static org.elasticsearch.index.IndexSettings.INDEX_MAPPING_SOURCE_EXCLUDE_SOURCE_VECTORS_SETTING;
 import static org.elasticsearch.index.seqno.SequenceNumbers.UNASSIGNED_PRIMARY_TERM;
 import static org.elasticsearch.index.seqno.SequenceNumbers.UNASSIGNED_SEQ_NO;
 
@@ -417,7 +417,7 @@ private GetResult innerGetFetch(
      */
     public static boolean shouldExcludeVectorsFromSource(IndexSettings indexSettings, FetchSourceContext fetchSourceContext) {
         if (fetchSourceContext == null || fetchSourceContext.excludeVectors() == null) {
-            return INDEX_MAPPING_SOURCE_SYNTHETIC_VECTORS_SETTING.get(indexSettings.getSettings());
+            return INDEX_MAPPING_SOURCE_EXCLUDE_SOURCE_VECTORS_SETTING.get(indexSettings.getSettings());
         }
         return fetchSourceContext.excludeVectors();
     }
diff --git a/server/src/main/java/org/elasticsearch/index/mapper/DocumentParser.java b/server/src/main/java/org/elasticsearch/index/mapper/DocumentParser.java
index 15e7ff88350b6..05a137b42e2fe 100644
--- a/server/src/main/java/org/elasticsearch/index/mapper/DocumentParser.java
+++ b/server/src/main/java/org/elasticsearch/index/mapper/DocumentParser.java
@@ -806,7 +806,7 @@ private static void postProcessDynamicArrayMapping(DocumentParserContext context
             DenseVectorFieldMapper.Builder builder = new DenseVectorFieldMapper.Builder(
                 fieldName,
                 context.indexSettings().getIndexVersionCreated(),
-                IndexSettings.INDEX_MAPPING_SOURCE_SYNTHETIC_VECTORS_SETTING.get(context.indexSettings().getSettings())
+                IndexSettings.INDEX_MAPPING_SOURCE_EXCLUDE_SOURCE_VECTORS_SETTING.get(context.indexSettings().getSettings())
             );
             builder.dimensions(mappers.size());
             DenseVectorFieldMapper denseVectorFieldMapper = builder.build(builderContext);
diff --git a/server/src/main/java/org/elasticsearch/index/mapper/vectors/DenseVectorFieldMapper.java b/server/src/main/java/org/elasticsearch/index/mapper/vectors/DenseVectorFieldMapper.java
index 9019edc435eaf..a6cb968c0093d 100644
--- a/server/src/main/java/org/elasticsearch/index/mapper/vectors/DenseVectorFieldMapper.java
+++ b/server/src/main/java/org/elasticsearch/index/mapper/vectors/DenseVectorFieldMapper.java
@@ -117,7 +117,7 @@
 import static org.elasticsearch.cluster.metadata.IndexMetadata.SETTING_INDEX_VERSION_CREATED;
 import static org.elasticsearch.common.Strings.format;
 import static org.elasticsearch.common.xcontent.XContentParserUtils.ensureExpectedToken;
-import static org.elasticsearch.index.IndexSettings.INDEX_MAPPING_SOURCE_SYNTHETIC_VECTORS_SETTING;
+import static org.elasticsearch.index.IndexSettings.INDEX_MAPPING_SOURCE_EXCLUDE_SOURCE_VECTORS_SETTING;
 import static org.elasticsearch.index.codec.vectors.IVFVectorsFormat.MAX_VECTORS_PER_CLUSTER;
 import static org.elasticsearch.index.codec.vectors.IVFVectorsFormat.MIN_VECTORS_PER_CLUSTER;
 
@@ -255,9 +255,9 @@ public static class Builder extends FieldMapper.Builder {
         private final Parameter<Map<String, String>> meta = Parameter.metaParam();
 
         final IndexVersion indexVersionCreated;
-        final boolean isSyntheticVector;
+        final boolean isExcludeSourceVectors;
 
-        public Builder(String name, IndexVersion indexVersionCreated, boolean isSyntheticVector) {
+        public Builder(String name, IndexVersion indexVersionCreated, boolean isExcludeSourceVectors) {
             super(name);
             this.indexVersionCreated = indexVersionCreated;
             // This is defined as updatable because it can be updated once, from [null] to a valid dim size,
@@ -289,7 +289,7 @@ public Builder(String name, IndexVersion indexVersionCreated, boolean isSyntheti
                         }
                     }
                 });
-            this.isSyntheticVector = isSyntheticVector;
+            this.isExcludeSourceVectors = isExcludeSourceVectors;
             final boolean indexedByDefault = indexVersionCreated.onOrAfter(INDEXED_BY_DEFAULT_INDEX_VERSION);
             final boolean defaultInt8Hnsw = indexVersionCreated.onOrAfter(IndexVersions.DEFAULT_DENSE_VECTOR_TO_INT8_HNSW);
             final boolean defaultBBQ8Hnsw = indexVersionCreated.onOrAfter(IndexVersions.DEFAULT_DENSE_VECTOR_TO_BBQ_HNSW);
@@ -431,7 +431,7 @@ public DenseVectorFieldMapper build(MapperBuilderContext context) {
             // Validate again here because the dimensions or element type could have been set programmatically,
             // which affects index option validity
             validate();
-            boolean isSyntheticVectorFinal = (context.isSourceSynthetic() == false) && indexed.getValue() && isSyntheticVector;
+            boolean isExcludeSourceVectorsFinal = context.isSourceSynthetic() == false && indexed.getValue() && isExcludeSourceVectors;
             return new DenseVectorFieldMapper(
                 leafName(),
                 new DenseVectorFieldType(
@@ -448,7 +448,7 @@ public DenseVectorFieldMapper build(MapperBuilderContext context) {
                 builderParams(this, context),
                 indexOptions.getValue(),
                 indexVersionCreated,
-                isSyntheticVectorFinal
+                isExcludeSourceVectorsFinal
             );
         }
     }
@@ -2391,7 +2391,7 @@ public XContentBuilder toXContent(XContentBuilder builder, Params params) throws
         (n, c) -> new Builder(
             n,
             c.getIndexSettings().getIndexVersionCreated(),
-            INDEX_MAPPING_SOURCE_SYNTHETIC_VECTORS_SETTING.get(c.getIndexSettings().getSettings())
+            INDEX_MAPPING_SOURCE_EXCLUDE_SOURCE_VECTORS_SETTING.get(c.getIndexSettings().getSettings())
         ),
         notInMultiFields(CONTENT_TYPE)
     );
@@ -2849,7 +2849,7 @@ public List<Object> fetchValues(Source source, int doc, List<Object> ignoredValu
 
     private final DenseVectorIndexOptions indexOptions;
     private final IndexVersion indexCreatedVersion;
-    private final boolean isSyntheticVector;
+    private final boolean isExcludeSourceVectors;
 
     private DenseVectorFieldMapper(
         String simpleName,
@@ -2857,12 +2857,12 @@ private DenseVectorFieldMapper(
         BuilderParams params,
         DenseVectorIndexOptions indexOptions,
         IndexVersion indexCreatedVersion,
-        boolean isSyntheticVector
+        boolean isExcludeSourceVectorsFinal
     ) {
         super(simpleName, mappedFieldType, params);
         this.indexOptions = indexOptions;
         this.indexCreatedVersion = indexCreatedVersion;
-        this.isSyntheticVector = isSyntheticVector;
+        this.isExcludeSourceVectors = isExcludeSourceVectorsFinal;
     }
 
     @Override
@@ -2984,7 +2984,7 @@ protected String contentType() {
 
     @Override
     public FieldMapper.Builder getMergeBuilder() {
-        return new Builder(leafName(), indexCreatedVersion, isSyntheticVector).init(this);
+        return new Builder(leafName(), indexCreatedVersion, isExcludeSourceVectors).init(this);
     }
 
     private static DenseVectorIndexOptions parseIndexOptions(String fieldName, Object propNode, IndexVersion indexVersion) {
@@ -3040,7 +3040,7 @@ public String toString() {
 
     @Override
     public SourceLoader.SyntheticVectorsLoader syntheticVectorsLoader() {
-        if (isSyntheticVector) {
+        if (isExcludeSourceVectors) {
             var syntheticField = new IndexedSyntheticFieldLoader(indexCreatedVersion, fieldType().similarity);
             return new SyntheticVectorsPatchFieldLoader(syntheticField, syntheticField::copyVectorAsList);
         }
diff --git a/server/src/main/java/org/elasticsearch/index/mapper/vectors/SparseVectorFieldMapper.java b/server/src/main/java/org/elasticsearch/index/mapper/vectors/SparseVectorFieldMapper.java
index 8eacc68b45c88..5c9de81ab8b68 100644
--- a/server/src/main/java/org/elasticsearch/index/mapper/vectors/SparseVectorFieldMapper.java
+++ b/server/src/main/java/org/elasticsearch/index/mapper/vectors/SparseVectorFieldMapper.java
@@ -12,11 +12,8 @@
 import org.apache.lucene.document.FeatureField;
 import org.apache.lucene.index.IndexableField;
 import org.apache.lucene.index.LeafReader;
-import org.apache.lucene.index.LeafReaderContext;
 import org.apache.lucene.index.PostingsEnum;
-import org.apache.lucene.index.TermVectors;
 import org.apache.lucene.index.TermsEnum;
-import org.apache.lucene.search.DocIdSetIterator;
 import org.apache.lucene.search.MatchNoDocsQuery;
 import org.apache.lucene.search.Query;
 import org.apache.lucene.util.BytesRef;
@@ -42,8 +39,6 @@
 import org.elasticsearch.index.query.SearchExecutionContext;
 import org.elasticsearch.inference.WeightedToken;
 import org.elasticsearch.inference.WeightedTokensUtils;
-import org.elasticsearch.search.fetch.StoredFieldsSpec;
-import org.elasticsearch.search.lookup.Source;
 import org.elasticsearch.xcontent.ConstructingObjectParser;
 import org.elasticsearch.xcontent.DeprecationHandler;
 import org.elasticsearch.xcontent.NamedXContentRegistry;
@@ -63,7 +58,7 @@
 import java.util.Objects;
 import java.util.stream.Stream;
 
-import static org.elasticsearch.index.IndexSettings.INDEX_MAPPING_SOURCE_SYNTHETIC_VECTORS_SETTING;
+import static org.elasticsearch.index.IndexSettings.INDEX_MAPPING_SOURCE_EXCLUDE_SOURCE_VECTORS_SETTING;
 import static org.elasticsearch.index.query.AbstractQueryBuilder.DEFAULT_BOOST;
 import static org.elasticsearch.xcontent.ConstructingObjectParser.optionalConstructorArg;
 
@@ -95,8 +90,7 @@ private static SparseVectorFieldMapper toType(FieldMapper in) {
 
     public static class Builder extends FieldMapper.Builder {
         private final IndexVersion indexVersionCreated;
-
-        private final Parameter<Boolean> stored = Parameter.storeParam(m -> toType(m).fieldType().isStored(), false);
+        private final Parameter<Boolean> stored;
         private final Parameter<Map<String, String>> meta = Parameter.metaParam();
         private final Parameter<IndexOptions> indexOptions = new Parameter<>(
             SPARSE_VECTOR_INDEX_OPTIONS,
@@ -108,12 +102,13 @@ public static class Builder extends FieldMapper.Builder {
             Objects::toString
         ).acceptsNull().setSerializerCheck(this::indexOptionsSerializerCheck);
 
-        private boolean isSyntheticVector;
+        private final boolean isExcludeSourceVectors;
 
-        public Builder(String name, IndexVersion indexVersionCreated, boolean isSyntheticVector) {
+        public Builder(String name, IndexVersion indexVersionCreated, boolean isExcludeSourceVectors) {
             super(name);
+            this.stored = Parameter.boolParam("store", false, m -> toType(m).fieldType().isStored(), () -> isExcludeSourceVectors);
             this.indexVersionCreated = indexVersionCreated;
-            this.isSyntheticVector = isSyntheticVector;
+            this.isExcludeSourceVectors = isExcludeSourceVectors;
         }
 
         public Builder setStored(boolean value) {
@@ -133,19 +128,18 @@ public SparseVectorFieldMapper build(MapperBuilderContext context) {
                 builderIndexOptions = getDefaultIndexOptions(indexVersionCreated);
             }
 
-            final boolean syntheticVectorFinal = context.isSourceSynthetic() == false && isSyntheticVector;
-            final boolean storedFinal = stored.getValue() || syntheticVectorFinal;
+            final boolean isExcludeSourceVectorsFinal = isExcludeSourceVectors && context.isSourceSynthetic() == false && stored.get();
             return new SparseVectorFieldMapper(
                 leafName(),
                 new SparseVectorFieldType(
                     indexVersionCreated,
                     context.buildFullName(leafName()),
-                    storedFinal,
+                    stored.get(),
                     meta.getValue(),
                     builderIndexOptions
                 ),
                 builderParams(this, context),
-                syntheticVectorFinal
+                isExcludeSourceVectorsFinal
             );
         }
 
@@ -206,7 +200,7 @@ private static SparseVectorFieldMapper.IndexOptions parseIndexOptions(MappingPar
         return new Builder(
             n,
             c.indexVersionCreated(),
-            INDEX_MAPPING_SOURCE_SYNTHETIC_VECTORS_SETTING.get(c.getIndexSettings().getSettings())
+            INDEX_MAPPING_SOURCE_EXCLUDE_SOURCE_VECTORS_SETTING.get(c.getIndexSettings().getSettings())
         );
     }, notInMultiFields(CONTENT_TYPE));
 
@@ -251,9 +245,6 @@ public IndexFieldData.Builder fielddataBuilder(FieldDataContext fieldDataContext
 
         @Override
         public ValueFetcher valueFetcher(SearchExecutionContext context, String format) {
-            if (isStored()) {
-                return new SparseVectorValueFetcher(name());
-            }
             return SourceValueFetcher.identity(name(), context, format);
         }
 
@@ -313,16 +304,17 @@ private static String indexedValueForSearch(Object value) {
         }
     }
 
-    private final boolean isSyntheticVector;
+    private final boolean isExcludeSourceVectors;
 
     private SparseVectorFieldMapper(
         String simpleName,
         MappedFieldType mappedFieldType,
         BuilderParams builderParams,
-        boolean isSyntheticVector
+        boolean isExcludeSourceVectors
     ) {
         super(simpleName, mappedFieldType, builderParams);
-        this.isSyntheticVector = isSyntheticVector;
+        assert isExcludeSourceVectors == false || fieldType().isStored();
+        this.isExcludeSourceVectors = isExcludeSourceVectors;
     }
 
     @Override
@@ -335,7 +327,7 @@ protected SyntheticSourceSupport syntheticSourceSupport() {
 
     @Override
     public SourceLoader.SyntheticVectorsLoader syntheticVectorsLoader() {
-        if (isSyntheticVector) {
+        if (isExcludeSourceVectors) {
             var syntheticField = new SparseVectorSyntheticFieldLoader(fullPath(), leafName());
             return new SyntheticVectorsPatchFieldLoader(syntheticField, syntheticField::copyAsMap);
         }
@@ -349,7 +341,7 @@ public Map<String, NamedAnalyzer> indexAnalyzers() {
 
     @Override
     public FieldMapper.Builder getMergeBuilder() {
-        return new Builder(leafName(), this.fieldType().indexVersionCreated, this.isSyntheticVector).init(this);
+        return new Builder(leafName(), this.fieldType().indexVersionCreated, this.isExcludeSourceVectors).init(this);
     }
 
     @Override
@@ -433,51 +425,6 @@ private static boolean indexVersionSupportsDefaultPruningConfig(IndexVersion ind
             || indexVersion.between(SPARSE_VECTOR_PRUNING_INDEX_OPTIONS_VERSION_8_X, IndexVersions.UPGRADE_TO_LUCENE_10_0_0));
     }
 
-    private static class SparseVectorValueFetcher implements ValueFetcher {
-        private final String fieldName;
-        private TermVectors termVectors;
-
-        private SparseVectorValueFetcher(String fieldName) {
-            this.fieldName = fieldName;
-        }
-
-        @Override
-        public void setNextReader(LeafReaderContext context) {
-            try {
-                termVectors = context.reader().termVectors();
-            } catch (IOException exc) {
-                throw new UncheckedIOException(exc);
-            }
-        }
-
-        @Override
-        public List<Object> fetchValues(Source source, int doc, List<Object> ignoredValues) throws IOException {
-            if (termVectors == null) {
-                return List.of();
-            }
-            var terms = termVectors.get(doc, fieldName);
-            if (terms == null) {
-                return List.of();
-            }
-
-            var termsEnum = terms.iterator();
-            PostingsEnum postingsScratch = null;
-            Map<String, Float> result = new LinkedHashMap<>();
-            while (termsEnum.next() != null) {
-                postingsScratch = termsEnum.postings(postingsScratch);
-                postingsScratch.nextDoc();
-                result.put(termsEnum.term().utf8ToString(), XFeatureField.decodeFeatureValue(postingsScratch.freq()));
-                assert postingsScratch.nextDoc() == DocIdSetIterator.NO_MORE_DOCS;
-            }
-            return List.of(result);
-        }
-
-        @Override
-        public StoredFieldsSpec storedFieldsSpec() {
-            return StoredFieldsSpec.NO_REQUIREMENTS;
-        }
-    }
-
     private static class SparseVectorSyntheticFieldLoader implements SourceLoader.SyntheticFieldLoader {
         private final String fullPath;
         private final String leafName;
diff --git a/server/src/main/java/org/elasticsearch/rest/action/search/SearchCapabilities.java b/server/src/main/java/org/elasticsearch/rest/action/search/SearchCapabilities.java
index 66648b7126514..17bd106a225e7 100644
--- a/server/src/main/java/org/elasticsearch/rest/action/search/SearchCapabilities.java
+++ b/server/src/main/java/org/elasticsearch/rest/action/search/SearchCapabilities.java
@@ -12,8 +12,6 @@
 import java.util.HashSet;
 import java.util.Set;
 
-import static org.elasticsearch.index.IndexSettings.SYNTHETIC_VECTORS;
-
 /**
  * A {@link Set} of "capabilities" supported by the {@link RestSearchAction}.
  */
@@ -55,8 +53,8 @@ private SearchCapabilities() {}
     private static final String EXCLUDE_VECTORS_PARAM = "exclude_vectors_param";
     private static final String DENSE_VECTOR_UPDATABLE_BBQ = "dense_vector_updatable_bbq";
     private static final String FIELD_EXISTS_QUERY_FOR_TEXT_FIELDS_NO_INDEX_OR_DV = "field_exists_query_for_text_fields_no_index_or_dv";
-    private static final String SYNTHETIC_VECTORS_SETTING = "synthetic_vectors_setting";
     private static final String UPDATE_FIELD_TO_BBQ_DISK = "update_field_to_bbq_disk";
+    private static final String EXCLUDE_SOURCE_VECTORS_SETTING = "exclude_source_vectors_setting";
 
     public static final Set<String> CAPABILITIES;
     static {
@@ -82,9 +80,7 @@ private SearchCapabilities() {}
         capabilities.add(DENSE_VECTOR_UPDATABLE_BBQ);
         capabilities.add(FIELD_EXISTS_QUERY_FOR_TEXT_FIELDS_NO_INDEX_OR_DV);
         capabilities.add(UPDATE_FIELD_TO_BBQ_DISK);
-        if (SYNTHETIC_VECTORS) {
-            capabilities.add(SYNTHETIC_VECTORS_SETTING);
-        }
+        capabilities.add(EXCLUDE_SOURCE_VECTORS_SETTING);
         CAPABILITIES = Set.copyOf(capabilities);
     }
 }
diff --git a/server/src/test/java/org/elasticsearch/index/mapper/vectors/DenseVectorFieldMapperTests.java b/server/src/test/java/org/elasticsearch/index/mapper/vectors/DenseVectorFieldMapperTests.java
index a5c5a9a9b42ef..05181e603586b 100644
--- a/server/src/test/java/org/elasticsearch/index/mapper/vectors/DenseVectorFieldMapperTests.java
+++ b/server/src/test/java/org/elasticsearch/index/mapper/vectors/DenseVectorFieldMapperTests.java
@@ -2462,22 +2462,7 @@ protected Object generateRandomInputValue(MappedFieldType ft) {
         DenseVectorFieldType vectorFieldType = (DenseVectorFieldType) ft;
         return switch (vectorFieldType.getElementType()) {
             case BYTE -> randomByteArrayOfLength(vectorFieldType.getVectorDimensions());
-            case FLOAT -> {
-                float[] floats = new float[vectorFieldType.getVectorDimensions()];
-                float magnitude = 0;
-                for (int i = 0; i < floats.length; i++) {
-                    float f = randomFloat();
-                    floats[i] = f;
-                    magnitude += f * f;
-                }
-                magnitude = (float) Math.sqrt(magnitude);
-                if (VectorSimilarity.DOT_PRODUCT.equals(vectorFieldType.getSimilarity())) {
-                    for (int i = 0; i < floats.length; i++) {
-                        floats[i] /= magnitude;
-                    }
-                }
-                yield floats;
-            }
+            case FLOAT -> randomNormalizedVector(vectorFieldType.getVectorDimensions());
             case BIT -> randomByteArrayOfLength(vectorFieldType.getVectorDimensions() / 8);
         };
     }
diff --git a/server/src/test/java/org/elasticsearch/index/mapper/vectors/SparseVectorFieldMapperTests.java b/server/src/test/java/org/elasticsearch/index/mapper/vectors/SparseVectorFieldMapperTests.java
index dc21e5f8f57cc..afb9f03903050 100644
--- a/server/src/test/java/org/elasticsearch/index/mapper/vectors/SparseVectorFieldMapperTests.java
+++ b/server/src/test/java/org/elasticsearch/index/mapper/vectors/SparseVectorFieldMapperTests.java
@@ -98,7 +98,7 @@ protected void minimalFieldMappingPreviousIndexDefaultsIncluded(XContentBuilder
 
     protected void minimalMappingWithExplicitDefaults(XContentBuilder b) throws IOException {
         b.field("type", "sparse_vector");
-        b.field("store", false);
+        b.field("store", true);
 
         b.startObject("meta");
         b.endObject();
@@ -394,7 +394,7 @@ public void testStoreIsNotUpdateable() throws IOException {
             .startObject("properties")
             .startObject("field")
             .field("type", "sparse_vector")
-            .field("store", true)
+            .field("store", false)
             .endObject()
             .endObject()
             .endObject()
@@ -447,23 +447,16 @@ protected boolean allowsNullValues() {
 
     @Override
     protected SyntheticSourceSupport syntheticSourceSupport(boolean syntheticSource) {
-        boolean withStore = randomBoolean();
         return new SyntheticSourceSupport() {
             @Override
             public boolean preservesExactSource() {
-                return withStore == false;
+                return false;
             }
 
             @Override
             public SyntheticSourceExample example(int maxValues) {
                 var sample = getSampleValueForDocument();
-                return new SyntheticSourceExample(sample, sample, b -> {
-                    if (withStore) {
-                        minimalStoreMapping(b);
-                    } else {
-                        minimalMapping(b);
-                    }
-                });
+                return new SyntheticSourceExample(sample, sample, b -> { minimalMapping(b); });
             }
 
             @Override
diff --git a/server/src/test/java/org/elasticsearch/index/mapper/vectors/SyntheticVectorFieldsRecoveryTests.java b/server/src/test/java/org/elasticsearch/index/mapper/vectors/SyntheticVectorFieldsRecoveryTests.java
index 138d138b741e5..e5cb5c89ad0eb 100644
--- a/server/src/test/java/org/elasticsearch/index/mapper/vectors/SyntheticVectorFieldsRecoveryTests.java
+++ b/server/src/test/java/org/elasticsearch/index/mapper/vectors/SyntheticVectorFieldsRecoveryTests.java
@@ -38,7 +38,6 @@
 import java.util.ArrayList;
 import java.util.List;
 
-import static org.elasticsearch.index.IndexSettings.SYNTHETIC_VECTORS;
 import static org.hamcrest.Matchers.equalTo;
 
 public class SyntheticVectorFieldsRecoveryTests extends EngineTestCase {
@@ -69,7 +68,7 @@ protected Settings indexSettings() {
             builder.put(IndexSettings.INDEX_MAPPER_SOURCE_MODE_SETTING.getKey(), SourceFieldMapper.Mode.SYNTHETIC.name());
             builder.put(IndexSettings.RECOVERY_USE_SYNTHETIC_SOURCE_SETTING.getKey(), useSyntheticRecovery);
         }
-        builder.put(IndexSettings.INDEX_MAPPING_SOURCE_SYNTHETIC_VECTORS_SETTING.getKey(), true);
+        builder.put(IndexSettings.INDEX_MAPPING_SOURCE_EXCLUDE_SOURCE_VECTORS_SETTING.getKey(), true);
         return builder.build();
     }
 
@@ -113,7 +112,6 @@ protected String defaultMapping() {
     }
 
     public void testSnapshotRecovery() throws IOException {
-        assumeTrue("feature flag must be enabled for synthetic vectors", SYNTHETIC_VECTORS);
         List<Translog.Index> expectedOperations = new ArrayList<>();
         int size = randomIntBetween(10, 50);
         for (int i = 0; i < size; i++) {
diff --git a/server/src/test/java/org/elasticsearch/index/mapper/vectors/SyntheticVectorsMapperTestCase.java b/server/src/test/java/org/elasticsearch/index/mapper/vectors/SyntheticVectorsMapperTestCase.java
index ebb4fe788fea3..d26710c2b6a03 100644
--- a/server/src/test/java/org/elasticsearch/index/mapper/vectors/SyntheticVectorsMapperTestCase.java
+++ b/server/src/test/java/org/elasticsearch/index/mapper/vectors/SyntheticVectorsMapperTestCase.java
@@ -26,12 +26,10 @@
 
 import java.io.IOException;
 
-import static org.elasticsearch.index.IndexSettings.SYNTHETIC_VECTORS;
 import static org.elasticsearch.test.hamcrest.ElasticsearchAssertions.assertToXContentEquivalent;
 
 public abstract class SyntheticVectorsMapperTestCase extends MapperTestCase {
     public void testSyntheticVectorsMinimalValidDocument() throws IOException {
-        assumeTrue("feature flag must be enabled for synthetic vectors", SYNTHETIC_VECTORS);
         for (XContentType type : XContentType.values()) {
             BytesReference source = generateRandomDoc(type, true, true, false, false, false);
             assertSyntheticVectors(buildVectorMapping(), source, type);
@@ -39,7 +37,6 @@ public void testSyntheticVectorsMinimalValidDocument() throws IOException {
     }
 
     public void testSyntheticVectorsFullDocument() throws IOException {
-        assumeTrue("feature flag must be enabled for synthetic vectors", SYNTHETIC_VECTORS);
         for (XContentType type : XContentType.values()) {
             BytesReference source = generateRandomDoc(type, true, true, true, true, false);
             assertSyntheticVectors(buildVectorMapping(), source, type);
@@ -47,7 +44,6 @@ public void testSyntheticVectorsFullDocument() throws IOException {
     }
 
     public void testSyntheticVectorsWithUnmappedFields() throws IOException {
-        assumeTrue("feature flag must be enabled for synthetic vectors", SYNTHETIC_VECTORS);
         for (XContentType type : XContentType.values()) {
             BytesReference source = generateRandomDoc(type, true, true, true, true, true);
             assertSyntheticVectors(buildVectorMapping(), source, type);
@@ -55,7 +51,6 @@ public void testSyntheticVectorsWithUnmappedFields() throws IOException {
     }
 
     public void testSyntheticVectorsMissingRootFields() throws IOException {
-        assumeTrue("feature flag must be enabled for synthetic vectors", SYNTHETIC_VECTORS);
         for (XContentType type : XContentType.values()) {
             BytesReference source = generateRandomDoc(type, false, false, false, false, false);
             assertSyntheticVectors(buildVectorMapping(), source, type);
@@ -63,7 +58,6 @@ public void testSyntheticVectorsMissingRootFields() throws IOException {
     }
 
     public void testSyntheticVectorsPartialNestedContent() throws IOException {
-        assumeTrue("feature flag must be enabled for synthetic vectors", SYNTHETIC_VECTORS);
         for (XContentType type : XContentType.values()) {
             BytesReference source = generateRandomDoc(type, true, true, true, false, false);
             assertSyntheticVectors(buildVectorMapping(), source, type);
@@ -71,7 +65,6 @@ public void testSyntheticVectorsPartialNestedContent() throws IOException {
     }
 
     public void testFlatPathDocument() throws IOException {
-        assumeTrue("feature flag must be enabled for synthetic vectors", SYNTHETIC_VECTORS);
         for (XContentType type : XContentType.values()) {
             BytesReference source = generateRandomDocWithFlatPath(type);
             assertSyntheticVectors(buildVectorMapping(), source, type);
@@ -248,7 +241,7 @@ private BytesReference generateRandomDocWithFlatPath(XContentType xContentType)
     }
 
     private void assertSyntheticVectors(String mapping, BytesReference source, XContentType xContentType) throws IOException {
-        var settings = Settings.builder().put(IndexSettings.INDEX_MAPPING_SOURCE_SYNTHETIC_VECTORS_SETTING.getKey(), true).build();
+        var settings = Settings.builder().put(IndexSettings.INDEX_MAPPING_SOURCE_EXCLUDE_SOURCE_VECTORS_SETTING.getKey(), true).build();
         MapperService mapperService = createMapperService(settings, mapping);
         var parsedDoc = mapperService.documentMapper().parse(new SourceToParse("0", source, xContentType));
         try (var directory = newDirectory()) {
diff --git a/server/src/test/java/org/elasticsearch/index/shard/ShardGetServiceTests.java b/server/src/test/java/org/elasticsearch/index/shard/ShardGetServiceTests.java
index f38b6ef660eff..554612c87d49a 100644
--- a/server/src/test/java/org/elasticsearch/index/shard/ShardGetServiceTests.java
+++ b/server/src/test/java/org/elasticsearch/index/shard/ShardGetServiceTests.java
@@ -36,7 +36,6 @@
 import java.util.Arrays;
 import java.util.function.LongSupplier;
 
-import static org.elasticsearch.index.IndexSettings.SYNTHETIC_VECTORS;
 import static org.elasticsearch.index.seqno.SequenceNumbers.UNASSIGNED_PRIMARY_TERM;
 import static org.elasticsearch.index.seqno.SequenceNumbers.UNASSIGNED_SEQ_NO;
 import static org.hamcrest.Matchers.equalTo;
@@ -138,11 +137,19 @@ public void testGetFromTranslogWithDenseVector() throws IOException {
                 "foo": "foo"
             }
             """, Arrays.toString(vector));
-        runGetFromTranslogWithOptions(docToIndex, "\"enabled\": true", null, docToIndex, "\"text\"", "foo", "\"dense_vector\"", false);
+        runGetFromTranslogWithOptions(
+            docToIndex,
+            "\"enabled\": true",
+            Settings.builder().put(IndexSettings.INDEX_MAPPING_SOURCE_EXCLUDE_SOURCE_VECTORS_SETTING.getKey(), false).build(),
+            docToIndex,
+            "\"text\"",
+            "foo",
+            "\"dense_vector\"",
+            false
+        );
     }
 
     public void testGetFromTranslogWithSyntheticVector() throws IOException {
-        assumeTrue("feature flag must be enabled for synthetic vectors", SYNTHETIC_VECTORS);
         float[] vector = new float[2048];
         for (int i = 0; i < vector.length; i++) {
             vector[i] = randomByte();
@@ -156,7 +163,7 @@ public void testGetFromTranslogWithSyntheticVector() throws IOException {
         runGetFromTranslogWithOptions(
             docToIndex,
             "\"enabled\": true",
-            Settings.builder().put(IndexSettings.INDEX_MAPPING_SOURCE_SYNTHETIC_VECTORS_SETTING.getKey(), true).build(),
+            Settings.builder().put(IndexSettings.INDEX_MAPPING_SOURCE_EXCLUDE_SOURCE_VECTORS_SETTING.getKey(), true).build(),
             docToIndex,
             "\"text\"",
             "foo",
diff --git a/x-pack/plugin/rank-vectors/src/main/java/org/elasticsearch/xpack/rank/vectors/RankVectorsPlugin.java b/x-pack/plugin/rank-vectors/src/main/java/org/elasticsearch/xpack/rank/vectors/RankVectorsPlugin.java
index dd38367125692..06663007519da 100644
--- a/x-pack/plugin/rank-vectors/src/main/java/org/elasticsearch/xpack/rank/vectors/RankVectorsPlugin.java
+++ b/x-pack/plugin/rank-vectors/src/main/java/org/elasticsearch/xpack/rank/vectors/RankVectorsPlugin.java
@@ -20,7 +20,7 @@
 
 import java.util.Map;
 
-import static org.elasticsearch.index.IndexSettings.INDEX_MAPPING_SOURCE_SYNTHETIC_VECTORS_SETTING;
+import static org.elasticsearch.index.IndexSettings.INDEX_MAPPING_SOURCE_EXCLUDE_SOURCE_VECTORS_SETTING;
 import static org.elasticsearch.index.mapper.FieldMapper.notInMultiFields;
 import static org.elasticsearch.xpack.rank.vectors.mapper.RankVectorsFieldMapper.CONTENT_TYPE;
 
@@ -41,7 +41,7 @@ public Map<String, Mapper.TypeParser> getMappers() {
                 n,
                 c.indexVersionCreated(),
                 getLicenseState(),
-                INDEX_MAPPING_SOURCE_SYNTHETIC_VECTORS_SETTING.get(c.getIndexSettings().getSettings())
+                INDEX_MAPPING_SOURCE_EXCLUDE_SOURCE_VECTORS_SETTING.get(c.getIndexSettings().getSettings())
             );
         }, notInMultiFields(CONTENT_TYPE)));
     }
diff --git a/x-pack/plugin/rank-vectors/src/main/java/org/elasticsearch/xpack/rank/vectors/mapper/RankVectorsFieldMapper.java b/x-pack/plugin/rank-vectors/src/main/java/org/elasticsearch/xpack/rank/vectors/mapper/RankVectorsFieldMapper.java
index a79fb4f304f6a..e2f314abc553f 100644
--- a/x-pack/plugin/rank-vectors/src/main/java/org/elasticsearch/xpack/rank/vectors/mapper/RankVectorsFieldMapper.java
+++ b/x-pack/plugin/rank-vectors/src/main/java/org/elasticsearch/xpack/rank/vectors/mapper/RankVectorsFieldMapper.java
@@ -113,13 +113,13 @@ public static class Builder extends FieldMapper.Builder {
 
         private final IndexVersion indexCreatedVersion;
         private final XPackLicenseState licenseState;
-        private final boolean isSyntheticVector;
+        private final boolean isExcludeSourceVectors;
 
-        public Builder(String name, IndexVersion indexCreatedVersion, XPackLicenseState licenseState, boolean isSyntheticVector) {
+        public Builder(String name, IndexVersion indexCreatedVersion, XPackLicenseState licenseState, boolean isExcludeSourceVectors) {
             super(name);
             this.indexCreatedVersion = indexCreatedVersion;
             this.licenseState = licenseState;
-            this.isSyntheticVector = isSyntheticVector;
+            this.isExcludeSourceVectors = isExcludeSourceVectors;
         }
 
         public Builder dimensions(int dimensions) {
@@ -141,7 +141,7 @@ public RankVectorsFieldMapper build(MapperBuilderContext context) {
             // Validate again here because the dimensions or element type could have been set programmatically,
             // which affects index option validity
             validate();
-            boolean isSyntheticVectorFinal = context.isSourceSynthetic() == false && isSyntheticVector;
+            boolean isExcludeSourceVectorsFinal = context.isSourceSynthetic() == false && isExcludeSourceVectors;
             return new RankVectorsFieldMapper(
                 leafName(),
                 new RankVectorsFieldType(
@@ -154,7 +154,7 @@ public RankVectorsFieldMapper build(MapperBuilderContext context) {
                 builderParams(this, context),
                 indexCreatedVersion,
                 licenseState,
-                isSyntheticVectorFinal
+                isExcludeSourceVectorsFinal
             );
         }
     }
@@ -252,7 +252,7 @@ DenseVectorFieldMapper.ElementType getElementType() {
 
     private final IndexVersion indexCreatedVersion;
     private final XPackLicenseState licenseState;
-    private final boolean isSyntheticVector;
+    private final boolean isExcludeSourceVectors;
 
     private RankVectorsFieldMapper(
         String simpleName,
@@ -260,12 +260,12 @@ private RankVectorsFieldMapper(
         BuilderParams params,
         IndexVersion indexCreatedVersion,
         XPackLicenseState licenseState,
-        boolean isSyntheticVector
+        boolean isExcludeSourceVectors
     ) {
         super(simpleName, fieldType, params);
         this.indexCreatedVersion = indexCreatedVersion;
         this.licenseState = licenseState;
-        this.isSyntheticVector = isSyntheticVector;
+        this.isExcludeSourceVectors = isExcludeSourceVectors;
     }
 
     @Override
@@ -396,7 +396,7 @@ protected String contentType() {
 
     @Override
     public FieldMapper.Builder getMergeBuilder() {
-        return new Builder(leafName(), indexCreatedVersion, licenseState, isSyntheticVector).init(this);
+        return new Builder(leafName(), indexCreatedVersion, licenseState, isExcludeSourceVectors).init(this);
     }
 
     @Override
@@ -406,7 +406,7 @@ protected SyntheticSourceSupport syntheticSourceSupport() {
 
     @Override
     public SourceLoader.SyntheticVectorsLoader syntheticVectorsLoader() {
-        if (isSyntheticVector) {
+        if (isExcludeSourceVectors) {
             var syntheticField = new DocValuesSyntheticFieldLoader();
             return new SyntheticVectorsPatchFieldLoader(syntheticField, syntheticField::copyVectorsAsList);
         }
diff --git a/x-pack/plugin/src/yamlRestTest/resources/rest-api-spec/test/ml/sparse_vector_search.yml b/x-pack/plugin/src/yamlRestTest/resources/rest-api-spec/test/ml/sparse_vector_search.yml
index 2d440f1cf5987..b5ee02bdbf875 100644
--- a/x-pack/plugin/src/yamlRestTest/resources/rest-api-spec/test/ml/sparse_vector_search.yml
+++ b/x-pack/plugin/src/yamlRestTest/resources/rest-api-spec/test/ml/sparse_vector_search.yml
@@ -372,6 +372,11 @@ teardown:
   - requires:
       cluster_features: [ "gte_v8.16.0" ]
       reason: dots in field names allowed starting in in 8.16.0
+      test_runner_features: [ capabilities, "close_to" ]
+      capabilities:
+        - method: GET
+          path: /_search
+          capabilities: [ exclude_source_vectors_setting ]
 
   - do:
       indices.create:
@@ -409,17 +414,14 @@ teardown:
       get:
         index: index-with-sparse-vector2
         id: "has-dots"
+        _source_exclude_vectors: false
 
-  - match:
-      _source:
-        ml:
-          tokens:
-            running: 2.4097164
-            good: 2.170997
-            run: 2.052153
-            race: 1.4575411
-            for: 1.1908325
-            5.0k: 2.489943
+  - close_to: { _source.ml.tokens.running: { value: 2.4097164, error: 0.01 } }
+  - close_to: { _source.ml.tokens.good: { value: 2.170997, error: 0.01 } }
+  - close_to: { _source.ml.tokens.run: { value: 2.052153, error: 0.01 } }
+  - close_to: { _source.ml.tokens.race: { value: 1.4575411, error: 0.01 } }
+  - close_to: { _source.ml.tokens.for: { value: 1.1908325, error: 0.01 } }
+  - close_to: { _source.ml.tokens.5\\.0k: { value: 2.489943, error: 0.01 } }
 
   - do:
       search:
@@ -439,6 +441,11 @@ teardown:
   - requires:
       cluster_features: [ "gte_v8.16.0" ]
       reason: dots in field names allowed starting in in 8.16.0
+      test_runner_features: [ capabilities, "close_to" ]
+      capabilities:
+        - method: GET
+          path: /_search
+          capabilities: [ exclude_source_vectors_setting ]
 
   - do:
       indices.create:
@@ -479,6 +486,7 @@ teardown:
       get:
         index: index-with-sparse-vector3
         id: "parent-foo"
+        _source_exclude_vectors: false
 
   - match:
       _source:
@@ -491,6 +499,7 @@ teardown:
       get:
         index: index-with-sparse-vector3
         id: "parent-foo-bar"
+        _source_exclude_vectors: false
 
   - match:
       _source:
diff --git a/x-pack/plugin/src/yamlRestTest/resources/rest-api-spec/test/rank_vectors/rank_vectors_synthetic_vectors.yml b/x-pack/plugin/src/yamlRestTest/resources/rest-api-spec/test/rank_vectors/rank_vectors_synthetic_vectors.yml
index c0df9d6a79d38..b39325d5147a8 100644
--- a/x-pack/plugin/src/yamlRestTest/resources/rest-api-spec/test/rank_vectors/rank_vectors_synthetic_vectors.yml
+++ b/x-pack/plugin/src/yamlRestTest/resources/rest-api-spec/test/rank_vectors/rank_vectors_synthetic_vectors.yml
@@ -6,7 +6,7 @@ setup:
       capabilities:
         - method: GET
           path: /_search
-          capabilities: [ synthetic_vectors_setting ]
+          capabilities: [ exclude_source_vectors_setting ]
   - skip:
       features: "headers"
 
@@ -14,8 +14,6 @@ setup:
       indices.create:
         index: test
         body:
-          settings:
-            index.mapping.synthetic_vectors: true
           mappings:
             properties:
               name:

From a8f12fef4f5b64ef42d1517c26ec3207973691e1 Mon Sep 17 00:00:00 2001
From: Jim Ferenczi <jim.ferenczi@elastic.co>
Date: Fri, 25 Jul 2025 10:12:24 +0200
Subject: [PATCH 02/16] Update docs/changelog/131907.yaml

---
 docs/changelog/131907.yaml | 12 ++++++++++++
 1 file changed, 12 insertions(+)
 create mode 100644 docs/changelog/131907.yaml

diff --git a/docs/changelog/131907.yaml b/docs/changelog/131907.yaml
new file mode 100644
index 0000000000000..a68bf780cf944
--- /dev/null
+++ b/docs/changelog/131907.yaml
@@ -0,0 +1,12 @@
+pr: 131907
+summary: Enable `exclude_source_vectors` by default for new indices
+area: Vector Search
+type: "breaking, enhancement"
+issues: []
+breaking:
+  title: Enable `exclude_source_vectors` by default for new indices
+  area: Vector Search
+  details: Please describe the details of this change for the release notes. You can
+    use asciidoc.
+  impact: Please describe the impact of this change to users
+  notable: false

From b0f932f7a85cb9ab29f930dc052215cf64653af8 Mon Sep 17 00:00:00 2001
From: Jim Ferenczi <jim.ferenczi@elastic.co>
Date: Fri, 25 Jul 2025 10:34:41 +0200
Subject: [PATCH 03/16] Update docs

---
 .../mapping-reference/dense-vector.md         | 16 +++++++--------
 .../mapping-reference/rank-vectors.md         | 18 ++++++++---------
 .../mapping-reference/sparse-vector.md        | 20 ++++++++++---------
 3 files changed, 28 insertions(+), 26 deletions(-)

diff --git a/docs/reference/elasticsearch/mapping-reference/dense-vector.md b/docs/reference/elasticsearch/mapping-reference/dense-vector.md
index 45dc923859c6b..8de654ffd1abb 100644
--- a/docs/reference/elasticsearch/mapping-reference/dense-vector.md
+++ b/docs/reference/elasticsearch/mapping-reference/dense-vector.md
@@ -102,7 +102,7 @@ This helps reduce response size and improve performance, especially in scenarios
 
 To retrieve vector values explicitly, you can use:
 
-* **The `fields` option** to request specific vector fields directly:
+* The `fields` option to request specific vector fields directly:
 
 ```console
 POST my-index-2/_search
@@ -111,7 +111,7 @@ POST my-index-2/_search
 }
 ```
 
-- **The `_source.exclude_vectors` flag** to re-enable vector inclusion in `_source` responses:
+- The `_source.exclude_vectors` flag to re-enable vector inclusion in `_source` responses:
 
 ```console
 POST my-index-2/_search
@@ -130,17 +130,17 @@ This setting is enabled by default for newly created indices and can only be set
 When enabled:
 
 * `dense_vector` fields are removed from `_source` and the rest of the `_source` is stored as usual.
-* If a request includes `_source` and vector values are needed (e.g., during recovery or reindex), the vectors are **rehydrated** from their internal format.
+* If a request includes `_source` and vector values are needed (e.g., during recovery or reindex), the vectors are rehydrated from their internal format.
 
-This setting is **compatible with synthetic `_source`**, where the entire `_source` document is reconstructed from columnar storage. In full synthetic mode, no `_source` is stored on disk, and all fields — including vectors — are rebuilt when needed.
+This setting is compatible with synthetic `_source`, where the entire `_source` document is reconstructed from columnar storage. In full synthetic mode, no `_source` is stored on disk, and all fields — including vectors — are rebuilt when needed.
 
 ### Rehydration and precision
 
-When vector values are rehydrated (e.g., for reindex, recovery, or explicit `_source` requests), they are restored from their internal format. Internally, vectors are stored at **float precision**, so if they were originally indexed as higher-precision types (e.g., `double` or `long`), the rehydrated values will have **reduced precision**. This lossy representation is intended to save space while preserving search quality.
+When vector values are rehydrated (e.g., for reindex, recovery, or explicit `_source` requests), they are restored from their internal format. Internally, vectors are stored at float precision, so if they were originally indexed as higher-precision types (e.g., `double` or `long`), the rehydrated values will have reduced precision. This lossy representation is intended to save space while preserving search quality.
 
 ### Storing original vectors in `_source`
 
-If you want to preserve the **original vector values exactly as they were provided**, you can re-enable vector storage in `_source`:
+If you want to preserve the original vector values exactly as they were provided, you can re-enable vector storage in `_source`:
 
 ```console
 PUT my-index-include-vectors
@@ -161,8 +161,8 @@ PUT my-index-include-vectors
 When this setting is disabled:
 
 * `dense_vector` fields are stored as part of the `_source`, exactly as indexed.
-* The index will store both the original `_source` value and the internal representation used for vector search, resulting in **increased storage usage**.
-* **Vectors are once again returned in `_source` by default** in all relevant APIs, with no need to use `exclude_vectors` or `fields`.
+* The index will store both the original `_source` value and the internal representation used for vector search, resulting in increased storage usage.
+* Vectors are once again returned in `_source` by default in all relevant APIs, with no need to use `exclude_vectors` or `fields`.
 
 This configuration is appropriate when full source fidelity is required, such as for auditing or round-tripping exact input values.
 
diff --git a/docs/reference/elasticsearch/mapping-reference/rank-vectors.md b/docs/reference/elasticsearch/mapping-reference/rank-vectors.md
index 7490cee0760bf..13326c5144eba 100644
--- a/docs/reference/elasticsearch/mapping-reference/rank-vectors.md
+++ b/docs/reference/elasticsearch/mapping-reference/rank-vectors.md
@@ -115,7 +115,7 @@ This helps reduce response size and improve performance, especially in scenarios
 
 To retrieve vector values explicitly, you can use:
 
-* **The `fields` option** to request specific vector fields directly:
+* The `fields` option to request specific vector fields directly:
 
 ```console
 POST my-index-2/_search
@@ -124,7 +124,7 @@ POST my-index-2/_search
 }
 ```
 
-- **The `_source.exclude_vectors` flag** to re-enable vector inclusion in `_source` responses:
+- The `_source.exclude_vectors` flag to re-enable vector inclusion in `_source` responses:
 
 ```console
 POST my-index-2/_search
@@ -137,23 +137,23 @@ POST my-index-2/_search
 
 ### Storage behavior and `_source`
 
-By default, `rank_vectors` fields are **not stored in `_source`** on disk. This is also controlled by the index setting `index.mapping.exclude_source_vectors`.
+By default, `rank_vectors` fields are not stored in `_source` on disk. This is also controlled by the index setting `index.mapping.exclude_source_vectors`.
 This setting is enabled by default for newly created indices and can only be set at index creation time.
 
 When enabled:
 
 * `rank_vectors` fields are removed from `_source` and the rest of the `_source` is stored as usual.
-* If a request includes `_source` and vector values are needed (e.g., during recovery or reindex), the vectors are **rehydrated** from their internal format.
+* If a request includes `_source` and vector values are needed (e.g., during recovery or reindex), the vectors are rehydrated from their internal format.
 
-This setting is **compatible with synthetic `_source`**, where the entire `_source` document is reconstructed from columnar storage. In full synthetic mode, no `_source` is stored on disk, and all fields — including vectors — are rebuilt when needed.
+This setting is compatible with synthetic `_source`, where the entire `_source` document is reconstructed from columnar storage. In full synthetic mode, no `_source` is stored on disk, and all fields — including vectors — are rebuilt when needed.
 
 ### Rehydration and precision
 
-When vector values are rehydrated (e.g., for reindex, recovery, or explicit `_source` requests), they are restored from their internal format. Internally, vectors are stored at **float precision**, so if they were originally indexed as higher-precision types (e.g., `double` or `long`), the rehydrated values will have **reduced precision**. This lossy representation is intended to save space while preserving search quality.
+When vector values are rehydrated (e.g., for reindex, recovery, or explicit `_source` requests), they are restored from their internal format. Internally, vectors are stored at float precision, so if they were originally indexed as higher-precision types (e.g., `double` or `long`), the rehydrated values will have reduced precision. This lossy representation is intended to save space while preserving search quality.
 
 ### Storing original vectors in `_source`
 
-If you want to preserve the **original vector values exactly as they were provided**, you can re-enable vector storage in `_source`:
+If you want to preserve the original vector values exactly as they were provided, you can re-enable vector storage in `_source`:
 
 ```console
 PUT my-index-include-vectors
@@ -175,8 +175,8 @@ PUT my-index-include-vectors
 When this setting is disabled:
 
 * `rank_vectors` fields are stored as part of the `_source`, exactly as indexed.
-* The index will store both the original `_source` value and the internal representation used for vector search, resulting in **increased storage usage**.
-* **Vectors are once again returned in `_source` by default** in all relevant APIs, with no need to use `exclude_vectors` or `fields`.
+* The index will store both the original `_source` value and the internal representation used for vector search, resulting in increased storage usage.
+* Vectors are once again returned in `_source` by default in all relevant APIs, with no need to use `exclude_vectors` or `fields`.
 
 This configuration is appropriate when full source fidelity is required, such as for auditing or round-tripping exact input values.
 
diff --git a/docs/reference/elasticsearch/mapping-reference/sparse-vector.md b/docs/reference/elasticsearch/mapping-reference/sparse-vector.md
index 595f9a1df0992..a9ac2de76243f 100644
--- a/docs/reference/elasticsearch/mapping-reference/sparse-vector.md
+++ b/docs/reference/elasticsearch/mapping-reference/sparse-vector.md
@@ -97,7 +97,7 @@ This helps reduce response size and improve performance, especially in scenarios
 
 To retrieve vector values explicitly, you can use:
 
-* **The `fields` option** to request specific vector fields directly:
+* The `fields` option to request specific vector fields directly:
 
 ```console
 POST my-index-2/_search
@@ -106,7 +106,7 @@ POST my-index-2/_search
 }
 ```
 
-- **The `_source.exclude_vectors` flag** to re-enable vector inclusion in `_source` responses:
+- The `_source.exclude_vectors` flag to re-enable vector inclusion in `_source` responses:
 
 ```console
 POST my-index-2/_search
@@ -119,23 +119,25 @@ POST my-index-2/_search
 
 ### Storage behavior and `_source`
 
-By default, `sparse_vector` fields are **not stored in `_source`** on disk. This is also controlled by the index setting `index.mapping.exclude_source_vectors`.
+By default, `sparse_vector` fields are not stored in `_source` on disk. This is also controlled by the index setting `index.mapping.exclude_source_vectors`.
 This setting is enabled by default for newly created indices and can only be set at index creation time.
 
 When enabled:
 
 * `sparse_vector` fields are removed from `_source` and the rest of the `_source` is stored as usual.
-* If a request includes `_source` and vector values are needed (e.g., during recovery or reindex), the vectors are **rehydrated** from their internal format.
+* If a request includes `_source` and vector values are needed (e.g., during recovery or reindex), the vectors are rehydrated from their internal format.
 
-This setting is **compatible with synthetic `_source`**, where the entire `_source` document is reconstructed from columnar storage. In full synthetic mode, no `_source` is stored on disk, and all fields — including vectors — are rebuilt when needed.
+This setting is compatible with synthetic `_source`, where the entire `_source` document is reconstructed from columnar storage. In full synthetic mode, no `_source` is stored on disk, and all fields — including vectors — are rebuilt when needed.
 
 ### Rehydration and precision
 
-When vector values are rehydrated (e.g., for reindex, recovery, or explicit `_source` requests), they are restored from their internal format. Internally, vectors are stored at **float precision**, so if they were originally indexed as higher-precision types (e.g., `double` or `long`), the rehydrated values will have **reduced precision**. This lossy representation is intended to save space while preserving search quality.
+When vector values are rehydrated (e.g., for reindex, recovery, or explicit `_source` requests), they are restored from their internal format.
+Internally, vectors are stored as floats with 9 significant bits for the precision, so the rehydrated values will have reduced precision.
+This lossy representation is intended to save space while preserving search quality.
 
 ### Storing original vectors in `_source`
 
-If you want to preserve the **original vector values exactly as they were provided**, you can re-enable vector storage in `_source`:
+If you want to preserve the original vector values exactly as they were provided, you can re-enable vector storage in `_source`:
 
 ```console
 PUT my-index-include-vectors
@@ -156,8 +158,8 @@ PUT my-index-include-vectors
 When this setting is disabled:
 
 * `sparse_vector` fields are stored as part of the `_source`, exactly as indexed.
-* The index will store both the original `_source` value and the internal representation used for vector search, resulting in **increased storage usage**.
-* **Vectors are once again returned in `_source` by default** in all relevant APIs, with no need to use `exclude_vectors` or `fields`.
+* The index will store both the original `_source` value and the internal representation used for vector search, resulting in increased storage usage.
+* Vectors are once again returned in `_source` by default in all relevant APIs, with no need to use `exclude_vectors` or `fields`.
 
 This configuration is appropriate when full source fidelity is required, such as for auditing or round-tripping exact input values.
 

From a3d9dad34f5662a501b15afb788ec6784a416b3e Mon Sep 17 00:00:00 2001
From: Jim Ferenczi <jim.ferenczi@elastic.co>
Date: Fri, 25 Jul 2025 13:39:16 +0200
Subject: [PATCH 04/16] fix more tests

---
 rest-api-spec/build.gradle                    |  4 +++
 x-pack/plugin/build.gradle                    |  5 +++
 .../ml/integration/TextEmbeddingQueryIT.java  | 31 ++++++++++++++++++-
 .../ml/integration/TextExpansionQueryIT.java  |  3 ++
 ...rrf_retriever_search_api_compatibility.yml | 12 +++++--
 5 files changed, 51 insertions(+), 4 deletions(-)

diff --git a/rest-api-spec/build.gradle b/rest-api-spec/build.gradle
index 7b73575f76ef3..9161dbdb744dd 100644
--- a/rest-api-spec/build.gradle
+++ b/rest-api-spec/build.gradle
@@ -90,4 +90,8 @@ tasks.named("yamlRestCompatTestTransform").configure ({ task ->
   task.skipTest("indices.create/21_synthetic_source_stored/field param - keep root array", "Synthetic source keep arrays now stores leaf arrays natively")
   task.skipTest("cluster.info/30_info_thread_pool/Cluster HTTP Info", "The search_throttled thread pool has been removed")
   task.skipTest("synonyms/80_synonyms_from_index/Fail loading synonyms from index if synonyms_set doesn't exist", "Synonyms do no longer fail if the synonyms_set doesn't exist")
+  task.skipTest("get/100_synthetic_source/indexed dense vectors", "Vectors are not returned by default")
+  task.skipTest("get/100_synthetic_source/non-indexed dense vectors", "Vectors are not returned by default")
+  task.skipTest("search.vectors/90_sparse_vector/stored sparse_vector synthetic source", "Vectors are not returned by default")
+  task.skipTest("search.vectors/90_sparse_vector/sparse_vector synthetic source", "Vectors are not returned by default")
 })
diff --git a/x-pack/plugin/build.gradle b/x-pack/plugin/build.gradle
index f7d28c1ef8bfe..4fbc9fc85bf96 100644
--- a/x-pack/plugin/build.gradle
+++ b/x-pack/plugin/build.gradle
@@ -139,6 +139,11 @@ tasks.named("yamlRestCompatTestTransform").configure({ task ->
   task.skipTest("esql/192_lookup_join_on_aliases/alias-pattern-multiple", "Error message changed")
   task.skipTest("esql/192_lookup_join_on_aliases/fails when alias or pattern resolves to multiple", "Error message changed")
   task.skipTest("esql/10_basic/Test wrong LIMIT parameter", "Error message changed")
+  task.skipTest("ml/sparse_vector_search/Search on a sparse_vector field with dots in the field names")
+  task.skipTest("get/100_synthetic_source/indexed dense vectors", "Vectors are not returned by default")
+  task.skipTest("get/100_synthetic_source/non-indexed dense vectors", "Vectors are not returned by default")
+  task.skipTest("search.vectors/90_sparse_vector/stored sparse_vector synthetic source", "Vectors are not returned by default")
+  task.skipTest("search.vectors/90_sparse_vector/sparse_vector synthetic source", "Vectors are not returned by default")
 })
 
 tasks.named('yamlRestCompatTest').configure {
diff --git a/x-pack/plugin/ml/qa/native-multi-node-tests/src/javaRestTest/java/org/elasticsearch/xpack/ml/integration/TextEmbeddingQueryIT.java b/x-pack/plugin/ml/qa/native-multi-node-tests/src/javaRestTest/java/org/elasticsearch/xpack/ml/integration/TextEmbeddingQueryIT.java
index 620819a8898dd..e21fc7f9be1ab 100644
--- a/x-pack/plugin/ml/qa/native-multi-node-tests/src/javaRestTest/java/org/elasticsearch/xpack/ml/integration/TextEmbeddingQueryIT.java
+++ b/x-pack/plugin/ml/qa/native-multi-node-tests/src/javaRestTest/java/org/elasticsearch/xpack/ml/integration/TextEmbeddingQueryIT.java
@@ -21,6 +21,7 @@
 
 import static org.hamcrest.Matchers.closeTo;
 import static org.hamcrest.Matchers.containsString;
+import static org.hamcrest.Matchers.equalTo;
 import static org.hamcrest.Matchers.hasSize;
 
 /**
@@ -100,6 +101,9 @@ public class TextEmbeddingQueryIT extends PyTorchModelRestTestCase {
 
     private static final String TOP_LEVEL_KNN_TEMPLATE = """
         {
+          "_source": {
+            "exclude_vectors": false
+          },
           "knn": {
               "field": "%s",
               "k": 5,
@@ -114,6 +118,9 @@ public class TextEmbeddingQueryIT extends PyTorchModelRestTestCase {
         }""";
     private static final String TOP_LEVEL_KNN_FILTER_TEMPLATE = """
         {
+          "_source": {
+            "exclude_vectors": false
+          },
           "knn": {
               "field": "%s",
               "k": 5,
@@ -129,6 +136,9 @@ public class TextEmbeddingQueryIT extends PyTorchModelRestTestCase {
         }""";
     private static final String TOP_LEVEL_KNN_HYBRID_ALL = """
         {
+          "_source": {
+            "exclude_vectors": false
+          },
           "knn": {
               "field": "embedding",
               "k": 3,
@@ -146,6 +156,9 @@ public class TextEmbeddingQueryIT extends PyTorchModelRestTestCase {
         }""";
     private static final String TOP_LEVEL_KNN_HYBRID_MATCH = """
         {
+          "_source": {
+            "exclude_vectors": false
+          },
           "knn": {
               "field": "embedding",
               "k": 3,
@@ -163,6 +176,9 @@ public class TextEmbeddingQueryIT extends PyTorchModelRestTestCase {
 
     private static final String QUERY_DSL_KNN_TEMPLATE = """
         {
+          "_source": {
+            "exclude_vectors": false
+          },
           "query": {
               "knn" : {
                   "field": "%s",
@@ -178,6 +194,9 @@ public class TextEmbeddingQueryIT extends PyTorchModelRestTestCase {
         }""";
     private static final String QUERY_DSL_KNN_FILTER_TEMPLATE = """
         {
+          "_source": {
+            "exclude_vectors": false
+          },
           "query": {
               "knn" : {
                   "field": "%s",
@@ -194,6 +213,9 @@ public class TextEmbeddingQueryIT extends PyTorchModelRestTestCase {
         }""";
     private static final String QUERY_DSL_KNN_HYBRID_ALL = """
         {
+             "_source": {
+                "exclude_vectors": false
+             },
              "query": {
                  "bool": {
                      "should": [
@@ -220,6 +242,9 @@ public class TextEmbeddingQueryIT extends PyTorchModelRestTestCase {
          }""";
     private static final String QUERY_DSL_KNN_HYBRID_MATCH = """
         {
+             "_source": {
+                "exclude_vectors": false
+             },
              "query": {
                  "bool": {
                      "should": [
@@ -554,7 +579,11 @@ public void testModelWithPrefixStrings() throws IOException {
             // The top hit should have the search prefix
             assertEquals(searchPrefix + "my words", sourceText);
             List<Double> foundEmbedding = (List<Double>) MapHelper.dig("_source.embedding", topHit);
-            assertEquals(embeddings.get(0), foundEmbedding);
+            var expectedEmbeddings = embeddings.get(0);
+            assertThat(foundEmbedding.size(), equalTo(expectedEmbeddings.size()));
+            for (int i = 0; i < foundEmbedding.size(); i++) {
+                assertEquals(expectedEmbeddings.get(i), foundEmbedding.get(i), 0.01f);
+            }
         }
     }
 
diff --git a/x-pack/plugin/ml/qa/native-multi-node-tests/src/javaRestTest/java/org/elasticsearch/xpack/ml/integration/TextExpansionQueryIT.java b/x-pack/plugin/ml/qa/native-multi-node-tests/src/javaRestTest/java/org/elasticsearch/xpack/ml/integration/TextExpansionQueryIT.java
index f1e8c9a67df44..58108b2c70b38 100644
--- a/x-pack/plugin/ml/qa/native-multi-node-tests/src/javaRestTest/java/org/elasticsearch/xpack/ml/integration/TextExpansionQueryIT.java
+++ b/x-pack/plugin/ml/qa/native-multi-node-tests/src/javaRestTest/java/org/elasticsearch/xpack/ml/integration/TextExpansionQueryIT.java
@@ -276,6 +276,9 @@ protected Response textExpansionSearch(String index, String modelText, String mo
 
         request.setJsonEntity(Strings.format("""
             {
+                "_source": {
+                    "exclude_vectors": false
+                },
                 "query": {
                   "text_expansion": {
                     "%s": {
diff --git a/x-pack/plugin/rank-rrf/src/yamlRestTest/resources/rest-api-spec/test/rrf/700_rrf_retriever_search_api_compatibility.yml b/x-pack/plugin/rank-rrf/src/yamlRestTest/resources/rest-api-spec/test/rrf/700_rrf_retriever_search_api_compatibility.yml
index 01d645fbfb4f5..89d4cb74a6210 100644
--- a/x-pack/plugin/rank-rrf/src/yamlRestTest/resources/rest-api-spec/test/rrf/700_rrf_retriever_search_api_compatibility.yml
+++ b/x-pack/plugin/rank-rrf/src/yamlRestTest/resources/rest-api-spec/test/rrf/700_rrf_retriever_search_api_compatibility.yml
@@ -980,19 +980,25 @@ setup:
 
 ---
 "rrf retriever with inner_hits for sub-retriever":
+  - skip:
+      features: [ "headers" ]
   - requires:
       capabilities:
         - method: POST
           path: /_search
-          capabilities: [ nested_retriever_inner_hits_support ]
+          capabilities: [ nested_retriever_inner_hits_support, exclude_source_vectors_setting ]
       test_runner_features: capabilities
       reason: "Support for propagating nested retrievers' inner hits to the top-level compound retriever is required"
 
   - do:
+      headers:
+        # Force JSON content type so that we use a parser that interprets the floating-point score as a double
+        Content-Type: application/json
       search:
-        _source: false
         index: test
         body:
+          _source:
+            exclude_vectors: false
           retriever:
             rrf:
               retrievers: [
@@ -1058,7 +1064,7 @@ setup:
   - match: { hits.hits.0.inner_hits.nested_data_field.hits.total.value: 1 }
   - match: { hits.hits.0.inner_hits.nested_data_field.hits.hits.0.fields.nested_inner_hits.0.data.0: foo }
   - match: { hits.hits.0.inner_hits.nested_vector_field.hits.total.value: 1 }
-  - match: { hits.hits.0.inner_hits.nested_vector_field.hits.hits.0.fields.nested_inner_hits.0.paragraph_id: [ 1 ] }
+  - match: { hits.hits.0.inner_hits.nested_vector_field.hits.hits.0.fields.nested_inner_hits.0.paragraph_id: [ 1.0 ] }
 
   - match: { hits.hits.1.inner_hits.nested_data_field.hits.total.value: 3 }
   - match: { hits.hits.1.inner_hits.nested_data_field.hits.hits.0.fields.nested_inner_hits.0.data.0: bar }

From 93d3ab357889b5ab666d411607168822cb5d191d Mon Sep 17 00:00:00 2001
From: Jim Ferenczi <jim.ferenczi@elastic.co>
Date: Fri, 25 Jul 2025 13:58:20 +0200
Subject: [PATCH 05/16] remove non-applicable skip test

---
 x-pack/plugin/build.gradle | 4 ----
 1 file changed, 4 deletions(-)

diff --git a/x-pack/plugin/build.gradle b/x-pack/plugin/build.gradle
index 4fbc9fc85bf96..5ce74ed1fc0db 100644
--- a/x-pack/plugin/build.gradle
+++ b/x-pack/plugin/build.gradle
@@ -140,10 +140,6 @@ tasks.named("yamlRestCompatTestTransform").configure({ task ->
   task.skipTest("esql/192_lookup_join_on_aliases/fails when alias or pattern resolves to multiple", "Error message changed")
   task.skipTest("esql/10_basic/Test wrong LIMIT parameter", "Error message changed")
   task.skipTest("ml/sparse_vector_search/Search on a sparse_vector field with dots in the field names")
-  task.skipTest("get/100_synthetic_source/indexed dense vectors", "Vectors are not returned by default")
-  task.skipTest("get/100_synthetic_source/non-indexed dense vectors", "Vectors are not returned by default")
-  task.skipTest("search.vectors/90_sparse_vector/stored sparse_vector synthetic source", "Vectors are not returned by default")
-  task.skipTest("search.vectors/90_sparse_vector/sparse_vector synthetic source", "Vectors are not returned by default")
 })
 
 tasks.named('yamlRestCompatTest').configure {

From 1af3c2c4418f60888c6422ee17286ec93e81cb68 Mon Sep 17 00:00:00 2001
From: Jim Ferenczi <jim.ferenczi@elastic.co>
Date: Fri, 25 Jul 2025 14:10:28 +0200
Subject: [PATCH 06/16] remove non-applicable skip test (bis)

---
 x-pack/plugin/build.gradle | 1 -
 1 file changed, 1 deletion(-)

diff --git a/x-pack/plugin/build.gradle b/x-pack/plugin/build.gradle
index 5ce74ed1fc0db..f7d28c1ef8bfe 100644
--- a/x-pack/plugin/build.gradle
+++ b/x-pack/plugin/build.gradle
@@ -139,7 +139,6 @@ tasks.named("yamlRestCompatTestTransform").configure({ task ->
   task.skipTest("esql/192_lookup_join_on_aliases/alias-pattern-multiple", "Error message changed")
   task.skipTest("esql/192_lookup_join_on_aliases/fails when alias or pattern resolves to multiple", "Error message changed")
   task.skipTest("esql/10_basic/Test wrong LIMIT parameter", "Error message changed")
-  task.skipTest("ml/sparse_vector_search/Search on a sparse_vector field with dots in the field names")
 })
 
 tasks.named('yamlRestCompatTest').configure {

From cd3cec29817435db9fb0b2b581675bf0d1bab6f2 Mon Sep 17 00:00:00 2001
From: Jim Ferenczi <jim.ferenczi@elastic.co>
Date: Fri, 25 Jul 2025 16:57:39 +0200
Subject: [PATCH 07/16] skip tests

---
 x-pack/plugin/build.gradle | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/x-pack/plugin/build.gradle b/x-pack/plugin/build.gradle
index f7d28c1ef8bfe..9419db3cba8c7 100644
--- a/x-pack/plugin/build.gradle
+++ b/x-pack/plugin/build.gradle
@@ -139,6 +139,8 @@ tasks.named("yamlRestCompatTestTransform").configure({ task ->
   task.skipTest("esql/192_lookup_join_on_aliases/alias-pattern-multiple", "Error message changed")
   task.skipTest("esql/192_lookup_join_on_aliases/fails when alias or pattern resolves to multiple", "Error message changed")
   task.skipTest("esql/10_basic/Test wrong LIMIT parameter", "Error message changed")
+  task.skipTest("ml/sparse_vector_search/Search on a sparse_vector field with dots in the field names", "Vectors are no longer returned by default")
+  task.skipTest("ml/sparse_vector_search/Search on a nested sparse_vector field with dots in the field names and conflicting child fields", "Vectors are no longer returned by default")
 })
 
 tasks.named('yamlRestCompatTest').configure {

From 87c5d579ce6b8619ec388d7f8b95ea61b72346d6 Mon Sep 17 00:00:00 2001
From: Jim Ferenczi <jim.ferenczi@elastic.co>
Date: Fri, 25 Jul 2025 20:53:55 +0200
Subject: [PATCH 08/16] add capabilities

---
 x-pack/plugin/ml/qa/ml-with-security/build.gradle | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/x-pack/plugin/ml/qa/ml-with-security/build.gradle b/x-pack/plugin/ml/qa/ml-with-security/build.gradle
index d18f6da13cad2..97407b882651b 100644
--- a/x-pack/plugin/ml/qa/ml-with-security/build.gradle
+++ b/x-pack/plugin/ml/qa/ml-with-security/build.gradle
@@ -8,7 +8,7 @@ dependencies {
 // bring in machine learning rest test suite
 restResources {
   restApi {
-    include '_common', 'cluster', 'nodes', 'indices', 'index', 'search', 'get', 'count', 'ingest', 'bulk', 'ml', 'cat'
+    include '_common', 'capabilities', 'cluster', 'nodes', 'indices', 'index', 'search', 'get', 'count', 'ingest', 'bulk', 'ml', 'cat'
   }
   restTests {
     includeXpack 'ml'

From 96e2d276bee6bd1b76687036fd4dae1400e0809c Mon Sep 17 00:00:00 2001
From: Jim Ferenczi <jim.ferenczi@elastic.co>
Date: Fri, 25 Jul 2025 23:24:28 +0200
Subject: [PATCH 09/16] Fix the translog operation asserter to check for map
 equivalence rather than byte per byte if fields are synthetic

---
 .../engine/TranslogOperationAsserter.java     |  8 ++-
 .../index/translog/Translog.java              | 55 +++++++++++++++----
 2 files changed, 48 insertions(+), 15 deletions(-)

diff --git a/server/src/main/java/org/elasticsearch/index/engine/TranslogOperationAsserter.java b/server/src/main/java/org/elasticsearch/index/engine/TranslogOperationAsserter.java
index 2cd3a9f755ffb..0898711aee809 100644
--- a/server/src/main/java/org/elasticsearch/index/engine/TranslogOperationAsserter.java
+++ b/server/src/main/java/org/elasticsearch/index/engine/TranslogOperationAsserter.java
@@ -41,8 +41,10 @@ public boolean assertSameIndexOperation(Translog.Index o1, Translog.Index o2) th
                 if (engineConfig.getIndexSettings().isRecoverySourceSyntheticEnabled()
                     || engineConfig.getMapperService().mappingLookup().inferenceFields().isEmpty() == false
                     || engineConfig.getMapperService().mappingLookup().syntheticVectorFields().isEmpty() == false) {
-                    return super.assertSameIndexOperation(synthesizeSource(engineConfig, o1), o2)
-                        || super.assertSameIndexOperation(o1, synthesizeSource(engineConfig, o2));
+                    // for synthetic source and synthetic fields, we check that the resulting source map is equivalent
+                    // since ordering might not be preserved.
+                    return Translog.Index.equalsWithoutAutoGeneratedTimestamp(synthesizeSource(engineConfig, o1), o2, false)
+                        || Translog.Index.equalsWithoutAutoGeneratedTimestamp(o1, synthesizeSource(engineConfig, o2), false);
                 }
                 return false;
             }
@@ -99,6 +101,6 @@ static Translog.Snapshot newSnapshot(EngineConfig engineConfig, Translog.Index o
     }
 
     public boolean assertSameIndexOperation(Translog.Index o1, Translog.Index o2) throws IOException {
-        return Translog.Index.equalsWithoutAutoGeneratedTimestamp(o1, o2);
+        return Translog.Index.equalsWithoutAutoGeneratedTimestamp(o1, o2, true);
     }
 }
diff --git a/server/src/main/java/org/elasticsearch/index/translog/Translog.java b/server/src/main/java/org/elasticsearch/index/translog/Translog.java
index b1a203616b120..6e83a684cfa82 100644
--- a/server/src/main/java/org/elasticsearch/index/translog/Translog.java
+++ b/server/src/main/java/org/elasticsearch/index/translog/Translog.java
@@ -22,6 +22,7 @@
 import org.elasticsearch.common.io.stream.Writeable;
 import org.elasticsearch.common.lucene.uid.Versions;
 import org.elasticsearch.common.util.BigArrays;
+import org.elasticsearch.common.xcontent.XContentHelper;
 import org.elasticsearch.core.IOUtils;
 import org.elasticsearch.core.Nullable;
 import org.elasticsearch.core.Releasable;
@@ -35,6 +36,8 @@
 import org.elasticsearch.index.shard.AbstractIndexShardComponent;
 import org.elasticsearch.index.shard.IndexShardComponent;
 import org.elasticsearch.index.shard.ShardId;
+import org.elasticsearch.search.lookup.Source;
+import org.elasticsearch.xcontent.XContentParserConfiguration;
 
 import java.io.Closeable;
 import java.io.EOFException;
@@ -1226,9 +1229,9 @@ public Type opType() {
         @Override
         public long estimateSize() {
             return (2 * id.length()) + source.length() + (routing != null ? 2 * routing.length() : 0) + (4 * Long.BYTES); // timestamp,
-                                                                                                                          // seq_no,
-                                                                                                                          // primary_term,
-                                                                                                                          // and version
+            // seq_no,
+            // primary_term,
+            // and version
         }
 
         public String id() {
@@ -1275,7 +1278,7 @@ public boolean equals(Object o) {
             }
 
             Index other = (Index) o;
-            return autoGeneratedIdTimestamp == other.autoGeneratedIdTimestamp && equalsWithoutAutoGeneratedTimestamp(this, other);
+            return autoGeneratedIdTimestamp == other.autoGeneratedIdTimestamp && equalsWithoutAutoGeneratedTimestamp(this, other, true);
         }
 
         @Override
@@ -1311,15 +1314,43 @@ public long getAutoGeneratedIdTimestamp() {
             return autoGeneratedIdTimestamp;
         }
 
-        public static boolean equalsWithoutAutoGeneratedTimestamp(Translog.Index o1, Translog.Index o2) {
-            return o1.version == o2.version
-                && o1.seqNo == o2.seqNo
-                && o1.primaryTerm == o2.primaryTerm
-                && o1.id.equals(o2.id)
-                && o1.source.equals(o2.source)
-                && Objects.equals(o1.routing, o2.routing);
-        }
+        public static boolean equalsWithoutAutoGeneratedTimestamp(Translog.Index o1, Translog.Index o2, boolean checkSourceBytes) {
+            if (o1.version != o2.version
+                || o1.seqNo != o2.seqNo
+                || o1.primaryTerm != o2.primaryTerm
+                || o1.id.equals(o2.id) == false
+                || Objects.equals(o1.routing, o2.routing) == false) {
+                return false;
+            }
+
+            if (checkSourceBytes) {
+                return o1.source.equals(o2.source);
+            }
 
+            var s1 = Source.fromBytes(o1.source);
+            var s2 = Source.fromBytes(o2.source);
+            try (
+                var actualParser = XContentHelper.createParserNotCompressed(
+                    XContentParserConfiguration.EMPTY,
+                    s1.internalSourceRef(),
+                    s1.sourceContentType()
+                )
+            ) {
+                var actualMap = actualParser.map();
+                try (
+                    var expectedParser = XContentHelper.createParserNotCompressed(
+                        XContentParserConfiguration.EMPTY,
+                        s2.internalSourceRef(),
+                        s2.sourceContentType()
+                    )
+                ) {
+                    var expectedMap = expectedParser.map();
+                    return expectedMap.equals(actualMap);
+                }
+            } catch (IOException exc) {
+                return false;
+            }
+        }
     }
 
     public static final class Delete extends Operation {

From 12019cd35da26d36f2a20d50fb104b7d37e091a4 Mon Sep 17 00:00:00 2001
From: Jim Ferenczi <jim.ferenczi@elastic.co>
Date: Sat, 26 Jul 2025 00:08:26 +0200
Subject: [PATCH 10/16] naming

---
 .../java/org/elasticsearch/reindex/ReindexBasicTests.java     | 4 ++--
 .../org/elasticsearch/reindex/UpdateByQueryBasicTests.java    | 2 +-
 .../elasticsearch/common/settings/IndexScopedSettings.java    | 2 +-
 .../src/main/java/org/elasticsearch/index/IndexSettings.java  | 2 +-
 .../java/org/elasticsearch/index/get/ShardGetService.java     | 4 ++--
 .../java/org/elasticsearch/index/mapper/DocumentParser.java   | 2 +-
 .../index/mapper/vectors/DenseVectorFieldMapper.java          | 4 ++--
 .../index/mapper/vectors/SparseVectorFieldMapper.java         | 4 ++--
 .../mapper/vectors/SyntheticVectorFieldsRecoveryTests.java    | 2 +-
 .../index/mapper/vectors/SyntheticVectorsMapperTestCase.java  | 2 +-
 .../org/elasticsearch/index/shard/ShardGetServiceTests.java   | 4 ++--
 .../elasticsearch/xpack/rank/vectors/RankVectorsPlugin.java   | 4 ++--
 12 files changed, 18 insertions(+), 18 deletions(-)

diff --git a/modules/reindex/src/test/java/org/elasticsearch/reindex/ReindexBasicTests.java b/modules/reindex/src/test/java/org/elasticsearch/reindex/ReindexBasicTests.java
index e8f8efe8114d5..92aa897bf6287 100644
--- a/modules/reindex/src/test/java/org/elasticsearch/reindex/ReindexBasicTests.java
+++ b/modules/reindex/src/test/java/org/elasticsearch/reindex/ReindexBasicTests.java
@@ -182,12 +182,12 @@ public void testReindexFromComplexDateMathIndexName() throws Exception {
 
     public void testReindexIncludeVectors() throws Exception {
         var resp1 = prepareCreate("test").setSettings(
-            Settings.builder().put(IndexSettings.INDEX_MAPPING_SOURCE_EXCLUDE_SOURCE_VECTORS_SETTING.getKey(), true).build()
+            Settings.builder().put(IndexSettings.INDEX_MAPPING_EXCLUDE_SOURCE_VECTORS_SETTING.getKey(), true).build()
         ).setMapping("foo", "type=dense_vector,similarity=l2_norm", "bar", "type=sparse_vector").get();
         assertAcked(resp1);
 
         var resp2 = prepareCreate("test_reindex").setSettings(
-            Settings.builder().put(IndexSettings.INDEX_MAPPING_SOURCE_EXCLUDE_SOURCE_VECTORS_SETTING.getKey(), true).build()
+            Settings.builder().put(IndexSettings.INDEX_MAPPING_EXCLUDE_SOURCE_VECTORS_SETTING.getKey(), true).build()
         ).setMapping("foo", "type=dense_vector,similarity=l2_norm", "bar", "type=sparse_vector").get();
         assertAcked(resp2);
 
diff --git a/modules/reindex/src/test/java/org/elasticsearch/reindex/UpdateByQueryBasicTests.java b/modules/reindex/src/test/java/org/elasticsearch/reindex/UpdateByQueryBasicTests.java
index 2075b63862af6..2ff1a258afb1d 100644
--- a/modules/reindex/src/test/java/org/elasticsearch/reindex/UpdateByQueryBasicTests.java
+++ b/modules/reindex/src/test/java/org/elasticsearch/reindex/UpdateByQueryBasicTests.java
@@ -158,7 +158,7 @@ public void testMissingSources() {
 
     public void testUpdateByQueryIncludeVectors() throws Exception {
         var resp1 = prepareCreate("test").setSettings(
-            Settings.builder().put(IndexSettings.INDEX_MAPPING_SOURCE_EXCLUDE_SOURCE_VECTORS_SETTING.getKey(), true).build()
+            Settings.builder().put(IndexSettings.INDEX_MAPPING_EXCLUDE_SOURCE_VECTORS_SETTING.getKey(), true).build()
         ).setMapping("foo", "type=dense_vector,similarity=l2_norm", "bar", "type=sparse_vector").get();
         assertAcked(resp1);
 
diff --git a/server/src/main/java/org/elasticsearch/common/settings/IndexScopedSettings.java b/server/src/main/java/org/elasticsearch/common/settings/IndexScopedSettings.java
index db4278bf991d5..93ddb5d3fc485 100644
--- a/server/src/main/java/org/elasticsearch/common/settings/IndexScopedSettings.java
+++ b/server/src/main/java/org/elasticsearch/common/settings/IndexScopedSettings.java
@@ -241,7 +241,7 @@ public final class IndexScopedSettings extends AbstractScopedSettings {
         if (IndexSettings.DOC_VALUES_SKIPPER) {
             settings.add(IndexSettings.USE_DOC_VALUES_SKIPPER);
         }
-        settings.add(IndexSettings.INDEX_MAPPING_SOURCE_EXCLUDE_SOURCE_VECTORS_SETTING);
+        settings.add(IndexSettings.INDEX_MAPPING_EXCLUDE_SOURCE_VECTORS_SETTING);
         BUILT_IN_INDEX_SETTINGS = Collections.unmodifiableSet(settings);
     };
 
diff --git a/server/src/main/java/org/elasticsearch/index/IndexSettings.java b/server/src/main/java/org/elasticsearch/index/IndexSettings.java
index 3912e6297e293..53fb5cc8f75e1 100644
--- a/server/src/main/java/org/elasticsearch/index/IndexSettings.java
+++ b/server/src/main/java/org/elasticsearch/index/IndexSettings.java
@@ -848,7 +848,7 @@ private static String getIgnoreAboveDefaultValue(final Settings settings) {
         Property.Final
     );
 
-    public static final Setting<Boolean> INDEX_MAPPING_SOURCE_EXCLUDE_SOURCE_VECTORS_SETTING = Setting.boolSetting(
+    public static final Setting<Boolean> INDEX_MAPPING_EXCLUDE_SOURCE_VECTORS_SETTING = Setting.boolSetting(
         "index.mapping.exclude_source_vectors",
         settings -> String.valueOf(SETTING_INDEX_VERSION_CREATED.get(settings).onOrAfter(IndexVersions.EXCLUDE_SOURCE_VECTORS_DEFAULT)),
         Property.IndexScope,
diff --git a/server/src/main/java/org/elasticsearch/index/get/ShardGetService.java b/server/src/main/java/org/elasticsearch/index/get/ShardGetService.java
index ac6e6b0fea718..1786032abbb6f 100644
--- a/server/src/main/java/org/elasticsearch/index/get/ShardGetService.java
+++ b/server/src/main/java/org/elasticsearch/index/get/ShardGetService.java
@@ -60,7 +60,7 @@
 import java.util.function.Function;
 import java.util.stream.Collectors;
 
-import static org.elasticsearch.index.IndexSettings.INDEX_MAPPING_SOURCE_EXCLUDE_SOURCE_VECTORS_SETTING;
+import static org.elasticsearch.index.IndexSettings.INDEX_MAPPING_EXCLUDE_SOURCE_VECTORS_SETTING;
 import static org.elasticsearch.index.seqno.SequenceNumbers.UNASSIGNED_PRIMARY_TERM;
 import static org.elasticsearch.index.seqno.SequenceNumbers.UNASSIGNED_SEQ_NO;
 
@@ -417,7 +417,7 @@ private GetResult innerGetFetch(
      */
     public static boolean shouldExcludeVectorsFromSource(IndexSettings indexSettings, FetchSourceContext fetchSourceContext) {
         if (fetchSourceContext == null || fetchSourceContext.excludeVectors() == null) {
-            return INDEX_MAPPING_SOURCE_EXCLUDE_SOURCE_VECTORS_SETTING.get(indexSettings.getSettings());
+            return INDEX_MAPPING_EXCLUDE_SOURCE_VECTORS_SETTING.get(indexSettings.getSettings());
         }
         return fetchSourceContext.excludeVectors();
     }
diff --git a/server/src/main/java/org/elasticsearch/index/mapper/DocumentParser.java b/server/src/main/java/org/elasticsearch/index/mapper/DocumentParser.java
index 05a137b42e2fe..1004ef0b208c2 100644
--- a/server/src/main/java/org/elasticsearch/index/mapper/DocumentParser.java
+++ b/server/src/main/java/org/elasticsearch/index/mapper/DocumentParser.java
@@ -806,7 +806,7 @@ private static void postProcessDynamicArrayMapping(DocumentParserContext context
             DenseVectorFieldMapper.Builder builder = new DenseVectorFieldMapper.Builder(
                 fieldName,
                 context.indexSettings().getIndexVersionCreated(),
-                IndexSettings.INDEX_MAPPING_SOURCE_EXCLUDE_SOURCE_VECTORS_SETTING.get(context.indexSettings().getSettings())
+                IndexSettings.INDEX_MAPPING_EXCLUDE_SOURCE_VECTORS_SETTING.get(context.indexSettings().getSettings())
             );
             builder.dimensions(mappers.size());
             DenseVectorFieldMapper denseVectorFieldMapper = builder.build(builderContext);
diff --git a/server/src/main/java/org/elasticsearch/index/mapper/vectors/DenseVectorFieldMapper.java b/server/src/main/java/org/elasticsearch/index/mapper/vectors/DenseVectorFieldMapper.java
index a6cb968c0093d..6d617ffb6bd03 100644
--- a/server/src/main/java/org/elasticsearch/index/mapper/vectors/DenseVectorFieldMapper.java
+++ b/server/src/main/java/org/elasticsearch/index/mapper/vectors/DenseVectorFieldMapper.java
@@ -117,7 +117,7 @@
 import static org.elasticsearch.cluster.metadata.IndexMetadata.SETTING_INDEX_VERSION_CREATED;
 import static org.elasticsearch.common.Strings.format;
 import static org.elasticsearch.common.xcontent.XContentParserUtils.ensureExpectedToken;
-import static org.elasticsearch.index.IndexSettings.INDEX_MAPPING_SOURCE_EXCLUDE_SOURCE_VECTORS_SETTING;
+import static org.elasticsearch.index.IndexSettings.INDEX_MAPPING_EXCLUDE_SOURCE_VECTORS_SETTING;
 import static org.elasticsearch.index.codec.vectors.IVFVectorsFormat.MAX_VECTORS_PER_CLUSTER;
 import static org.elasticsearch.index.codec.vectors.IVFVectorsFormat.MIN_VECTORS_PER_CLUSTER;
 
@@ -2391,7 +2391,7 @@ public XContentBuilder toXContent(XContentBuilder builder, Params params) throws
         (n, c) -> new Builder(
             n,
             c.getIndexSettings().getIndexVersionCreated(),
-            INDEX_MAPPING_SOURCE_EXCLUDE_SOURCE_VECTORS_SETTING.get(c.getIndexSettings().getSettings())
+            INDEX_MAPPING_EXCLUDE_SOURCE_VECTORS_SETTING.get(c.getIndexSettings().getSettings())
         ),
         notInMultiFields(CONTENT_TYPE)
     );
diff --git a/server/src/main/java/org/elasticsearch/index/mapper/vectors/SparseVectorFieldMapper.java b/server/src/main/java/org/elasticsearch/index/mapper/vectors/SparseVectorFieldMapper.java
index 5c9de81ab8b68..5f25babffb3a0 100644
--- a/server/src/main/java/org/elasticsearch/index/mapper/vectors/SparseVectorFieldMapper.java
+++ b/server/src/main/java/org/elasticsearch/index/mapper/vectors/SparseVectorFieldMapper.java
@@ -58,7 +58,7 @@
 import java.util.Objects;
 import java.util.stream.Stream;
 
-import static org.elasticsearch.index.IndexSettings.INDEX_MAPPING_SOURCE_EXCLUDE_SOURCE_VECTORS_SETTING;
+import static org.elasticsearch.index.IndexSettings.INDEX_MAPPING_EXCLUDE_SOURCE_VECTORS_SETTING;
 import static org.elasticsearch.index.query.AbstractQueryBuilder.DEFAULT_BOOST;
 import static org.elasticsearch.xcontent.ConstructingObjectParser.optionalConstructorArg;
 
@@ -200,7 +200,7 @@ private static SparseVectorFieldMapper.IndexOptions parseIndexOptions(MappingPar
         return new Builder(
             n,
             c.indexVersionCreated(),
-            INDEX_MAPPING_SOURCE_EXCLUDE_SOURCE_VECTORS_SETTING.get(c.getIndexSettings().getSettings())
+            INDEX_MAPPING_EXCLUDE_SOURCE_VECTORS_SETTING.get(c.getIndexSettings().getSettings())
         );
     }, notInMultiFields(CONTENT_TYPE));
 
diff --git a/server/src/test/java/org/elasticsearch/index/mapper/vectors/SyntheticVectorFieldsRecoveryTests.java b/server/src/test/java/org/elasticsearch/index/mapper/vectors/SyntheticVectorFieldsRecoveryTests.java
index e5cb5c89ad0eb..3718dae1c7d3e 100644
--- a/server/src/test/java/org/elasticsearch/index/mapper/vectors/SyntheticVectorFieldsRecoveryTests.java
+++ b/server/src/test/java/org/elasticsearch/index/mapper/vectors/SyntheticVectorFieldsRecoveryTests.java
@@ -68,7 +68,7 @@ protected Settings indexSettings() {
             builder.put(IndexSettings.INDEX_MAPPER_SOURCE_MODE_SETTING.getKey(), SourceFieldMapper.Mode.SYNTHETIC.name());
             builder.put(IndexSettings.RECOVERY_USE_SYNTHETIC_SOURCE_SETTING.getKey(), useSyntheticRecovery);
         }
-        builder.put(IndexSettings.INDEX_MAPPING_SOURCE_EXCLUDE_SOURCE_VECTORS_SETTING.getKey(), true);
+        builder.put(IndexSettings.INDEX_MAPPING_EXCLUDE_SOURCE_VECTORS_SETTING.getKey(), true);
         return builder.build();
     }
 
diff --git a/server/src/test/java/org/elasticsearch/index/mapper/vectors/SyntheticVectorsMapperTestCase.java b/server/src/test/java/org/elasticsearch/index/mapper/vectors/SyntheticVectorsMapperTestCase.java
index d26710c2b6a03..f7a23383f4e92 100644
--- a/server/src/test/java/org/elasticsearch/index/mapper/vectors/SyntheticVectorsMapperTestCase.java
+++ b/server/src/test/java/org/elasticsearch/index/mapper/vectors/SyntheticVectorsMapperTestCase.java
@@ -241,7 +241,7 @@ private BytesReference generateRandomDocWithFlatPath(XContentType xContentType)
     }
 
     private void assertSyntheticVectors(String mapping, BytesReference source, XContentType xContentType) throws IOException {
-        var settings = Settings.builder().put(IndexSettings.INDEX_MAPPING_SOURCE_EXCLUDE_SOURCE_VECTORS_SETTING.getKey(), true).build();
+        var settings = Settings.builder().put(IndexSettings.INDEX_MAPPING_EXCLUDE_SOURCE_VECTORS_SETTING.getKey(), true).build();
         MapperService mapperService = createMapperService(settings, mapping);
         var parsedDoc = mapperService.documentMapper().parse(new SourceToParse("0", source, xContentType));
         try (var directory = newDirectory()) {
diff --git a/server/src/test/java/org/elasticsearch/index/shard/ShardGetServiceTests.java b/server/src/test/java/org/elasticsearch/index/shard/ShardGetServiceTests.java
index 554612c87d49a..6227a8cc221d5 100644
--- a/server/src/test/java/org/elasticsearch/index/shard/ShardGetServiceTests.java
+++ b/server/src/test/java/org/elasticsearch/index/shard/ShardGetServiceTests.java
@@ -140,7 +140,7 @@ public void testGetFromTranslogWithDenseVector() throws IOException {
         runGetFromTranslogWithOptions(
             docToIndex,
             "\"enabled\": true",
-            Settings.builder().put(IndexSettings.INDEX_MAPPING_SOURCE_EXCLUDE_SOURCE_VECTORS_SETTING.getKey(), false).build(),
+            Settings.builder().put(IndexSettings.INDEX_MAPPING_EXCLUDE_SOURCE_VECTORS_SETTING.getKey(), false).build(),
             docToIndex,
             "\"text\"",
             "foo",
@@ -163,7 +163,7 @@ public void testGetFromTranslogWithSyntheticVector() throws IOException {
         runGetFromTranslogWithOptions(
             docToIndex,
             "\"enabled\": true",
-            Settings.builder().put(IndexSettings.INDEX_MAPPING_SOURCE_EXCLUDE_SOURCE_VECTORS_SETTING.getKey(), true).build(),
+            Settings.builder().put(IndexSettings.INDEX_MAPPING_EXCLUDE_SOURCE_VECTORS_SETTING.getKey(), true).build(),
             docToIndex,
             "\"text\"",
             "foo",
diff --git a/x-pack/plugin/rank-vectors/src/main/java/org/elasticsearch/xpack/rank/vectors/RankVectorsPlugin.java b/x-pack/plugin/rank-vectors/src/main/java/org/elasticsearch/xpack/rank/vectors/RankVectorsPlugin.java
index 06663007519da..cf302a4bebe86 100644
--- a/x-pack/plugin/rank-vectors/src/main/java/org/elasticsearch/xpack/rank/vectors/RankVectorsPlugin.java
+++ b/x-pack/plugin/rank-vectors/src/main/java/org/elasticsearch/xpack/rank/vectors/RankVectorsPlugin.java
@@ -20,7 +20,7 @@
 
 import java.util.Map;
 
-import static org.elasticsearch.index.IndexSettings.INDEX_MAPPING_SOURCE_EXCLUDE_SOURCE_VECTORS_SETTING;
+import static org.elasticsearch.index.IndexSettings.INDEX_MAPPING_EXCLUDE_SOURCE_VECTORS_SETTING;
 import static org.elasticsearch.index.mapper.FieldMapper.notInMultiFields;
 import static org.elasticsearch.xpack.rank.vectors.mapper.RankVectorsFieldMapper.CONTENT_TYPE;
 
@@ -41,7 +41,7 @@ public Map<String, Mapper.TypeParser> getMappers() {
                 n,
                 c.indexVersionCreated(),
                 getLicenseState(),
-                INDEX_MAPPING_SOURCE_EXCLUDE_SOURCE_VECTORS_SETTING.get(c.getIndexSettings().getSettings())
+                INDEX_MAPPING_EXCLUDE_SOURCE_VECTORS_SETTING.get(c.getIndexSettings().getSettings())
             );
         }, notInMultiFields(CONTENT_TYPE)));
     }

From ff7bc4d730de91d3bb77c54871b4badaf4608807 Mon Sep 17 00:00:00 2001
From: Jim Ferenczi <jim.ferenczi@elastic.co>
Date: Mon, 28 Jul 2025 11:03:50 +0200
Subject: [PATCH 11/16] Update docs/changelog/131907.yaml

---
 docs/changelog/131907.yaml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/docs/changelog/131907.yaml b/docs/changelog/131907.yaml
index a68bf780cf944..f2537dc048480 100644
--- a/docs/changelog/131907.yaml
+++ b/docs/changelog/131907.yaml
@@ -1,7 +1,7 @@
 pr: 131907
 summary: Enable `exclude_source_vectors` by default for new indices
 area: Vector Search
-type: "breaking, enhancement"
+type: breaking
 issues: []
 breaking:
   title: Enable `exclude_source_vectors` by default for new indices

From 7c3f34a6d7b010907324ee4b739840aa307864af Mon Sep 17 00:00:00 2001
From: Jim Ferenczi <jim.ferenczi@elastic.co>
Date: Mon, 28 Jul 2025 10:08:28 +0100
Subject: [PATCH 12/16] Update change log

---
 docs/changelog/131907.yaml | 24 +++++++++++++++++++-----
 1 file changed, 19 insertions(+), 5 deletions(-)

diff --git a/docs/changelog/131907.yaml b/docs/changelog/131907.yaml
index f2537dc048480..5f6eaae296143 100644
--- a/docs/changelog/131907.yaml
+++ b/docs/changelog/131907.yaml
@@ -1,12 +1,26 @@
 pr: 131907
 summary: Enable `exclude_source_vectors` by default for new indices
 area: Vector Search
-type: breaking
+type: "breaking"
 issues: []
 breaking:
   title: Enable `exclude_source_vectors` by default for new indices
   area: Vector Search
-  details: Please describe the details of this change for the release notes. You can
-    use asciidoc.
-  impact: Please describe the impact of this change to users
-  notable: false
+  details: |
+    The `exclude_source_vectors` setting is now enabled by default for newly created indices.
+    This means that vector fields (e.g., `dense_vector`) are no longer stored in the `_source` field
+    by default, although they remain fully accessible through search and retrieval operations.
+
+    Instead of being persisted in `_source`, vectors are now rehydrated on demand from the underlying
+    index structures when needed. This reduces index size and improves performance for typical vector
+    search workloads where the original vector values do not need to be part of the `_source`.
+
+    If your use case requires vector fields to be stored in `_source`, you can disable this behavior by
+    setting `exclude_source_vectors: false` at index creation time.
+  impact: |
+    Vector fields will no longer be stored in `_source` by default for new indices. Applications or tools
+    that expect to see vector fields in `_source` (for raw document inspection)
+    may need to be updated or configured to explicitly retain vectors using `exclude_source_vectors: false`.
+
+    Retrieval of vector fields via search or the `_source` API remains fully supported.
+  notable: true

From 247f974de19131c3c12eda9311133fd2680938ac Mon Sep 17 00:00:00 2001
From: Jim Ferenczi <jim.ferenczi@elastic.co>
Date: Mon, 28 Jul 2025 10:21:55 +0100
Subject: [PATCH 13/16] Update change log

---
 docs/changelog/131907.yaml | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/docs/changelog/131907.yaml b/docs/changelog/131907.yaml
index 5f6eaae296143..210770be807dd 100644
--- a/docs/changelog/131907.yaml
+++ b/docs/changelog/131907.yaml
@@ -1,12 +1,12 @@
 pr: 131907
 summary: Enable `exclude_source_vectors` by default for new indices
 area: Vector Search
-type: "breaking"
+type: breaking
 issues: []
 breaking:
   title: Enable `exclude_source_vectors` by default for new indices
   area: Vector Search
-  details: |
+  details: |-
     The `exclude_source_vectors` setting is now enabled by default for newly created indices.
     This means that vector fields (e.g., `dense_vector`) are no longer stored in the `_source` field
     by default, although they remain fully accessible through search and retrieval operations.
@@ -17,7 +17,7 @@ breaking:
 
     If your use case requires vector fields to be stored in `_source`, you can disable this behavior by
     setting `exclude_source_vectors: false` at index creation time.
-  impact: |
+  impact: |-
     Vector fields will no longer be stored in `_source` by default for new indices. Applications or tools
     that expect to see vector fields in `_source` (for raw document inspection)
     may need to be updated or configured to explicitly retain vectors using `exclude_source_vectors: false`.

From 916ecfd503ac50b1f8c51174d7e4cfd7f0f76cf9 Mon Sep 17 00:00:00 2001
From: Jim Ferenczi <jim.ferenczi@elastic.co>
Date: Mon, 28 Jul 2025 11:20:35 +0100
Subject: [PATCH 14/16] changelog

---
 docs/changelog/131907.yaml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/docs/changelog/131907.yaml b/docs/changelog/131907.yaml
index 210770be807dd..6e8db5dbb272f 100644
--- a/docs/changelog/131907.yaml
+++ b/docs/changelog/131907.yaml
@@ -5,7 +5,7 @@ type: breaking
 issues: []
 breaking:
   title: Enable `exclude_source_vectors` by default for new indices
-  area: Vector Search
+  area: Search
   details: |-
     The `exclude_source_vectors` setting is now enabled by default for newly created indices.
     This means that vector fields (e.g., `dense_vector`) are no longer stored in the `_source` field

From 762e7be4214b9726dbd179400f9099ee88d76f4e Mon Sep 17 00:00:00 2001
From: Jim Ferenczi <jim.ferenczi@elastic.co>
Date: Mon, 4 Aug 2025 16:13:11 +0200
Subject: [PATCH 15/16] apply review comment

---
 .../mapping-reference/dense-vector.md         | 32 ++++++++++++++-----
 1 file changed, 24 insertions(+), 8 deletions(-)

diff --git a/docs/reference/elasticsearch/mapping-reference/dense-vector.md b/docs/reference/elasticsearch/mapping-reference/dense-vector.md
index 4f692ade30c08..1a78c7784a4b8 100644
--- a/docs/reference/elasticsearch/mapping-reference/dense-vector.md
+++ b/docs/reference/elasticsearch/mapping-reference/dense-vector.md
@@ -103,6 +103,10 @@ PUT my-index-2
 {{es}} uses the [HNSW algorithm](https://arxiv.org/abs/1603.09320) to support efficient kNN search. Like most kNN algorithms, HNSW is an approximate method that sacrifices result accuracy for improved speed.
 
 ## Accessing `dense_vector` fields in search responses
+```{applies_to}
+stack: ga 9.2
+serverless: ga
+```
 
 By default, `dense_vector` fields are **not included in `_source`** in responses from the `_search`, `_msearch`, `_get`, and `_mget` APIs.
 This helps reduce response size and improve performance, especially in scenarios where vectors are used solely for similarity scoring and not required in the output.
@@ -130,6 +134,10 @@ POST my-index-2/_search
 ```
 
 ### Storage behavior and `_source`
+```{applies_to}
+stack: ga 9.2
+serverless: ga
+```
 
 By default, `dense_vector` fields are **not stored in `_source`** on disk. This is also controlled by the index setting `index.mapping.exclude_source_vectors`.
 This setting is enabled by default for newly created indices and can only be set at index creation time.
@@ -142,10 +150,18 @@ When enabled:
 This setting is compatible with synthetic `_source`, where the entire `_source` document is reconstructed from columnar storage. In full synthetic mode, no `_source` is stored on disk, and all fields — including vectors — are rebuilt when needed.
 
 ### Rehydration and precision
+```{applies_to}
+stack: ga 9.2
+serverless: ga
+```
 
 When vector values are rehydrated (e.g., for reindex, recovery, or explicit `_source` requests), they are restored from their internal format. Internally, vectors are stored at float precision, so if they were originally indexed as higher-precision types (e.g., `double` or `long`), the rehydrated values will have reduced precision. This lossy representation is intended to save space while preserving search quality.
 
 ### Storing original vectors in `_source`
+```{applies_to}
+stack: ga 9.2
+serverless: ga
+```
 
 If you want to preserve the original vector values exactly as they were provided, you can re-enable vector storage in `_source`:
 
@@ -337,16 +353,16 @@ $$$dense-vector-index-options$$$
 `type`
 :   (Required, string) The type of kNN algorithm to use. Can be either any of:
     * `hnsw` - This utilizes the [HNSW algorithm](https://arxiv.org/abs/1603.09320) for scalable approximate kNN search. This supports all `element_type` values.
-    * `int8_hnsw` - The default index type for some float vectors: 
-        
-      * {applies_to}`stack: ga 9.1` Default for float vectors with less than 384 dimensions. 
+    * `int8_hnsw` - The default index type for some float vectors:
+
+      * {applies_to}`stack: ga 9.1` Default for float vectors with less than 384 dimensions.
       * {applies_to}`stack: ga 9.0` Default for float all vectors.
-      
+
       This utilizes the [HNSW algorithm](https://arxiv.org/abs/1603.09320) in addition to automatically scalar quantization for scalable approximate kNN search with `element_type` of `float`. This can reduce the memory footprint by 4x at the cost of some accuracy. See [Automatically quantize vectors for kNN search](#dense-vector-quantization).
     * `int4_hnsw` - This utilizes the [HNSW algorithm](https://arxiv.org/abs/1603.09320) in addition to automatically scalar quantization for scalable approximate kNN search with `element_type` of `float`. This can reduce the memory footprint by 8x at the cost of some accuracy. See [Automatically quantize vectors for kNN search](#dense-vector-quantization).
     * `bbq_hnsw` - This utilizes the [HNSW algorithm](https://arxiv.org/abs/1603.09320) in addition to automatically binary quantization for scalable approximate kNN search with `element_type` of `float`. This can reduce the memory footprint by 32x at the cost of accuracy. See [Automatically quantize vectors for kNN search](#dense-vector-quantization).
-        
-      {applies_to}`stack: ga 9.1` `bbq_hnsw` is the default index type for float vectors with greater than or equal to 384 dimensions. 
+
+      {applies_to}`stack: ga 9.1` `bbq_hnsw` is the default index type for float vectors with greater than or equal to 384 dimensions.
     * `flat` - This utilizes a brute-force search algorithm for exact kNN search. This supports all `element_type` values.
     * `int8_flat` - This utilizes a brute-force search algorithm in addition to automatically scalar quantization. Only supports `element_type` of `float`.
     * `int4_flat` - This utilizes a brute-force search algorithm in addition to automatically half-byte scalar quantization. Only supports `element_type` of `float`.
@@ -366,8 +382,8 @@ $$$dense-vector-index-options$$$
 :   (Optional, object) An optional section that configures automatic vector rescoring on knn queries for the given field. Only applicable to quantized index types.
 :::::{dropdown} Properties of rescore_vector
 `oversample`
-:   (required, float) The amount to oversample the search results by. This value should be one of the following: 
-    * Greater than `1.0` and less than `10.0` 
+:   (required, float) The amount to oversample the search results by. This value should be one of the following:
+    * Greater than `1.0` and less than `10.0`
     * Exactly `0` to indicate no oversampling and rescoring should occur {applies_to}`stack: ga 9.1`
     :   The higher the value, the more vectors will be gathered and rescored with the raw values per shard.
     :   In case a knn query specifies a `rescore_vector` parameter, the query `rescore_vector` parameter will be used instead.

From 1828921c05ef9858a9db9b1285cc11618d23f446 Mon Sep 17 00:00:00 2001
From: Jim Ferenczi <jim.ferenczi@elastic.co>
Date: Mon, 4 Aug 2025 16:19:47 +0200
Subject: [PATCH 16/16] update docs

---
 .../elasticsearch/mapping-reference/dense-vector.md  | 12 ------------
 .../elasticsearch/mapping-reference/rank-vectors.md  |  4 ++++
 .../elasticsearch/mapping-reference/sparse-vector.md |  4 ++++
 3 files changed, 8 insertions(+), 12 deletions(-)

diff --git a/docs/reference/elasticsearch/mapping-reference/dense-vector.md b/docs/reference/elasticsearch/mapping-reference/dense-vector.md
index 1a78c7784a4b8..0dc84889d4799 100644
--- a/docs/reference/elasticsearch/mapping-reference/dense-vector.md
+++ b/docs/reference/elasticsearch/mapping-reference/dense-vector.md
@@ -134,10 +134,6 @@ POST my-index-2/_search
 ```
 
 ### Storage behavior and `_source`
-```{applies_to}
-stack: ga 9.2
-serverless: ga
-```
 
 By default, `dense_vector` fields are **not stored in `_source`** on disk. This is also controlled by the index setting `index.mapping.exclude_source_vectors`.
 This setting is enabled by default for newly created indices and can only be set at index creation time.
@@ -150,18 +146,10 @@ When enabled:
 This setting is compatible with synthetic `_source`, where the entire `_source` document is reconstructed from columnar storage. In full synthetic mode, no `_source` is stored on disk, and all fields — including vectors — are rebuilt when needed.
 
 ### Rehydration and precision
-```{applies_to}
-stack: ga 9.2
-serverless: ga
-```
 
 When vector values are rehydrated (e.g., for reindex, recovery, or explicit `_source` requests), they are restored from their internal format. Internally, vectors are stored at float precision, so if they were originally indexed as higher-precision types (e.g., `double` or `long`), the rehydrated values will have reduced precision. This lossy representation is intended to save space while preserving search quality.
 
 ### Storing original vectors in `_source`
-```{applies_to}
-stack: ga 9.2
-serverless: ga
-```
 
 If you want to preserve the original vector values exactly as they were provided, you can re-enable vector storage in `_source`:
 
diff --git a/docs/reference/elasticsearch/mapping-reference/rank-vectors.md b/docs/reference/elasticsearch/mapping-reference/rank-vectors.md
index 13326c5144eba..2c22dd08ae1b0 100644
--- a/docs/reference/elasticsearch/mapping-reference/rank-vectors.md
+++ b/docs/reference/elasticsearch/mapping-reference/rank-vectors.md
@@ -109,6 +109,10 @@ $$$rank-vectors-element-type$$$
 :   (Optional, integer) Number of vector dimensions. Can’t exceed `4096`. If `dims` is not specified, it will be set to the length of the first vector added to the field.
 
 ## Accessing `dense_vector` fields in search responses
+```{applies_to}
+stack: ga 9.2
+serverless: ga
+```
 
 By default, `dense_vector` fields are **not included in `_source`** in responses from the `_search`, `_msearch`, `_get`, and `_mget` APIs.
 This helps reduce response size and improve performance, especially in scenarios where vectors are used solely for similarity scoring and not required in the output.
diff --git a/docs/reference/elasticsearch/mapping-reference/sparse-vector.md b/docs/reference/elasticsearch/mapping-reference/sparse-vector.md
index a9ac2de76243f..5cb009ddcf302 100644
--- a/docs/reference/elasticsearch/mapping-reference/sparse-vector.md
+++ b/docs/reference/elasticsearch/mapping-reference/sparse-vector.md
@@ -91,6 +91,10 @@ This ensures that:
 * Very infrequent tokens that may not have as high of a score are removed.
 
 ## Accessing `dense_vector` fields in search responses
+```{applies_to}
+stack: ga 9.2
+serverless: ga
+```
 
 By default, `dense_vector` fields are **not included in `_source`** in responses from the `_search`, `_msearch`, `_get`, and `_mget` APIs.
 This helps reduce response size and improve performance, especially in scenarios where vectors are used solely for similarity scoring and not required in the output.