@@ -21,7 +21,6 @@ online_store:
     port: ES_PORT
     user: ES_USERNAME
     password: ES_PASSWORD
-    vector_len: 512
     write_batch_size: 1000
 ```
 {% endcode %}
@@ -88,7 +87,7 @@ Currently, the indexing mapping in the ElasticSearch online store is configured
     "created_ts": {"type": "date"},
     "vector_value": {
         "type": "dense_vector",
-        "dims": config.online_store.vector_len,
+        "dims": vector_field_length,
         "index": "true",
         "similarity": config.online_store.similarity,
     },
 
@@ -31,7 +31,6 @@ online_store:
     sslcert_path: /path/to/client-cert.pem
     sslrootcert_path: /path/to/server-ca.pem
     vector_enabled: false
-    vector_len: 512
 ```
 {% endcode %}
 
@@ -67,7 +66,7 @@ To compare this set of functionality against other online stores, please see the
 The Postgres online store supports the use of [PGVector](https://github.com/pgvector/pgvector) for storing feature values.
 To enable PGVector, set `vector_enabled: true` in the online store configuration. 
 
-The `vector_len` parameter can be used to specify the length of the vector. The default value is 512.
+The `vector_length` parameter can be used to specify the length of the vector in the Field.
 
 Please make sure to follow the instructions in the repository, which, as the time of this writing, requires you to 
 run `CREATE EXTENSION vector;` in the database.
 
@@ -20,7 +20,6 @@ online_store:
     type: qdrant
     host: localhost
     port: 6333
-    vector_len: 384
     write_batch_size: 100
 ```
 
 
@@ -42,4 +42,7 @@ message FeatureSpecV2 {
 
     // Metric used for vector similarity search.
     string vector_search_metric = 6;
+
+    // Field indicating the vector length
+    int32 vector_length = 7;
 }
@@ -62,7 +62,6 @@
 from feast.feast_object import FeastObject
 from feast.feature_service import FeatureService
 from feast.feature_view import DUMMY_ENTITY, DUMMY_ENTITY_NAME, FeatureView
-from feast.field import Field
 from feast.inference import (
     update_data_sources_with_inferred_event_timestamp_col,
     update_feature_views_with_inferred_features_and_entities,
@@ -91,7 +90,7 @@
 from feast.stream_feature_view import StreamFeatureView
 from feast.transformation.pandas_transformation import PandasTransformation
 from feast.transformation.python_transformation import PythonTransformation
-from feast.utils import _utc_now
+from feast.utils import _get_feature_view_vector_field_metadata, _utc_now
 
 warnings.simplefilter("once", DeprecationWarning)
 
@@ -856,7 +855,6 @@ def apply(
         if not isinstance(objects, Iterable):
             objects = [objects]
         assert isinstance(objects, list)
-
         if not objects_to_delete:
             objects_to_delete = []
 
@@ -1555,6 +1553,18 @@ def _get_feature_view_and_df_for_online_write(
                 except Exception as _:
                     raise DataFrameSerializationError(df)
 
+        if feature_view.features[0].vector_index and df is not None:
+            fv_vector_feature_name = feature_view.features[0].name
+            df_vector_feature_index = df.columns.get_loc(fv_vector_feature_name)
+            if feature_view.features[0].vector_length != 0:
+                if (
+                    df.shape[df_vector_feature_index]
+                    > feature_view.features[0].vector_length
+                ):
+                    raise ValueError(
+                        f"The dataframe for {fv_vector_feature_name} column has {df.shape[1]} vectors which is greater than expected (i.e {feature_view.features[0].vector_length}) by feature view {feature_view.name}."
+                    )
+
         # # Apply transformations if this is an OnDemandFeatureView with write_to_online_store=True
         if (
             isinstance(feature_view, OnDemandFeatureView)
@@ -2502,16 +2512,3 @@ def _validate_data_sources(data_sources: List[DataSource]):
             raise DataSourceRepeatNamesException(case_insensitive_ds_name)
         else:
             ds_names.add(case_insensitive_ds_name)
-
-
-def _get_feature_view_vector_field_metadata(
-    feature_view: FeatureView,
-) -> Optional[Field]:
-    vector_fields = [field for field in feature_view.schema if field.vector_index]
-    if len(vector_fields) > 1:
-        raise ValueError(
-            f"Feature view {feature_view.name} has multiple vector fields. Only one vector field per feature view is supported."
-        )
-    if not vector_fields:
-        return None
-    return vector_fields[0]
@@ -33,6 +33,7 @@ class Field:
         description: A human-readable description.
         tags: User-defined metadata in dictionary form.
         vector_index: If set to True the field will be indexed for vector similarity search.
+        vector_length: The length of the vector if the vector index is set to True.
         vector_search_metric: The metric used for vector similarity search.
     """
 
@@ -41,6 +42,7 @@ class Field:
     description: str
     tags: Dict[str, str]
     vector_index: bool
+    vector_length: int
     vector_search_metric: Optional[str]
 
     def __init__(
@@ -51,6 +53,7 @@ def __init__(
         description: str = "",
         tags: Optional[Dict[str, str]] = None,
         vector_index: bool = False,
+        vector_length: int = 0,
         vector_search_metric: Optional[str] = None,
     ):
         """
@@ -69,6 +72,7 @@ def __init__(
         self.description = description
         self.tags = tags or {}
         self.vector_index = vector_index
+        self.vector_length = vector_length
         self.vector_search_metric = vector_search_metric
 
     def __eq__(self, other):
@@ -80,6 +84,7 @@ def __eq__(self, other):
             or self.dtype != other.dtype
             or self.description != other.description
             or self.tags != other.tags
+            or self.vector_length != other.vector_length
             # or self.vector_index != other.vector_index
             # or self.vector_search_metric != other.vector_search_metric
         ):
@@ -100,6 +105,7 @@ def __repr__(self):
             f"    description={self.description!r},\n"
             f"    tags={self.tags!r}\n"
             f"    vector_index={self.vector_index!r}\n"
+            f"    vector_length={self.vector_length!r}\n"
             f"    vector_search_metric={self.vector_search_metric!r}\n"
             f")"
         )
@@ -117,6 +123,7 @@ def to_proto(self) -> FieldProto:
             description=self.description,
             tags=self.tags,
             vector_index=self.vector_index,
+            vector_length=self.vector_length,
             vector_search_metric=vector_search_metric,
         )
 
@@ -131,12 +138,14 @@ def from_proto(cls, field_proto: FieldProto):
         value_type = ValueType(field_proto.value_type)
         vector_search_metric = getattr(field_proto, "vector_search_metric", "")
         vector_index = getattr(field_proto, "vector_index", False)
+        vector_length = getattr(field_proto, "vector_length", 0)
         return cls(
             name=field_proto.name,
             dtype=from_value_type(value_type=value_type),
             tags=dict(field_proto.tags),
             description=field_proto.description,
             vector_index=vector_index,
+            vector_length=vector_length,
             vector_search_metric=vector_search_metric,
         )
 
 
@@ -18,7 +18,11 @@
 from feast.protos.feast.types.EntityKey_pb2 import EntityKey as EntityKeyProto
 from feast.protos.feast.types.Value_pb2 import Value as ValueProto
 from feast.repo_config import FeastConfigBaseModel
-from feast.utils import _build_retrieve_online_document_record, to_naive_utc
+from feast.utils import (
+    _build_retrieve_online_document_record,
+    _get_feature_view_vector_field_metadata,
+    to_naive_utc,
+)
 
 
 class ElasticSearchOnlineStoreConfig(FeastConfigBaseModel, VectorStoreConfig):
@@ -161,6 +165,10 @@ def create_index(self, config: RepoConfig, table: FeatureView):
             config: Feast repo configuration object.
             table: FeatureView table for which the index needs to be created.
         """
+        vector_field_length = getattr(
+            _get_feature_view_vector_field_metadata(table), "vector_length", 512
+        )
+
         index_mapping = {
             "properties": {
                 "entity_key": {"type": "binary"},
@@ -170,7 +178,7 @@ def create_index(self, config: RepoConfig, table: FeatureView):
                 "created_ts": {"type": "date"},
                 "vector_value": {
                     "type": "dense_vector",
-                    "dims": config.online_store.vector_len,
+                    "dims": vector_field_length,
                     "index": "true",
                     "similarity": config.online_store.similarity,
                 },
 
@@ -311,7 +311,7 @@ def update(
             for table in tables_to_keep:
                 table_name = _table_id(project, table)
                 if config.online_store.vector_enabled:
-                    vector_value_type = f"vector({config.online_store.vector_len})"
+                    vector_value_type = "vector"
                 else:
                     # keep the vector_value_type as BYTEA if pgvector is not enabled, to maintain compatibility
                     vector_value_type = "BYTEA"
 
@@ -19,7 +19,11 @@
 from feast.protos.feast.types.EntityKey_pb2 import EntityKey as EntityKeyProto
 from feast.protos.feast.types.Value_pb2 import Value as ValueProto
 from feast.repo_config import FeastConfigBaseModel
-from feast.utils import _build_retrieve_online_document_record, to_naive_utc
+from feast.utils import (
+    _build_retrieve_online_document_record,
+    _get_feature_view_vector_field_metadata,
+    to_naive_utc,
+)
 
 SCROLL_SIZE = 1000
 
@@ -198,13 +202,17 @@ def create_collection(self, config: RepoConfig, table: FeatureView):
             table: FeatureView table for which the index needs to be created.
         """
 
+        vector_field_length = getattr(
+            _get_feature_view_vector_field_metadata(table), "vector_length", 512
+        )
+
         client: QdrantClient = self._get_client(config)
 
         client.create_collection(
             collection_name=table.name,
             vectors_config={
                 config.online_store.vector_name: models.VectorParams(
-                    size=config.online_store.vector_len,
+                    size=vector_field_length,
                     distance=DISTANCE_MAPPING[config.online_store.similarity.lower()],
                 )
             },
 
@@ -43,6 +43,7 @@
 from feast.types import FEAST_VECTOR_TYPES, PrimitiveFeastType
 from feast.utils import (
     _build_retrieve_online_document_record,
+    _get_feature_view_vector_field_metadata,
     _serialize_vector_to_float_list,
     to_naive_utc,
 )
@@ -100,7 +101,6 @@ class SqliteOnlineStoreConfig(FeastConfigBaseModel, VectorStoreConfig):
     """ (optional) Path to sqlite db """
 
     vector_enabled: bool = False
-    vector_len: Optional[int] = None
     text_search_enabled: bool = False
 
 
@@ -171,8 +171,13 @@ def online_write_batch(
                             feature_type_dict.get(feature_name, None)
                             in FEAST_VECTOR_TYPES
                         ):
+                            vector_field_length = getattr(
+                                _get_feature_view_vector_field_metadata(table),
+                                "vector_length",
+                                512,
+                            )
                             val_bin = serialize_f32(
-                                val.float_list_val.val, config.online_store.vector_len
+                                val.float_list_val.val, vector_field_length
                             )  # type: ignore
                         else:
                             val_bin = feast_value_type_to_python_type(val)
@@ -354,15 +359,19 @@ def retrieve_online_documents(
         conn = self._get_conn(config)
         cur = conn.cursor()
 
+        vector_field_length = getattr(
+            _get_feature_view_vector_field_metadata(table), "vector_length", 512
+        )
+
         # Convert the embedding to a binary format instead of using SerializeToString()
-        query_embedding_bin = serialize_f32(embedding, config.online_store.vector_len)
+        query_embedding_bin = serialize_f32(embedding, vector_field_length)
         table_name = _table_id(project, table)
         vector_field = _get_vector_field(table)
 
         cur.execute(
             f"""
             CREATE VIRTUAL TABLE vec_table using vec0(
-                vector_value float[{config.online_store.vector_len}]
+                vector_value float[{vector_field_length}]
         );
         """
         )
@@ -378,7 +387,7 @@ def retrieve_online_documents(
         cur.execute(
             f"""
             CREATE VIRTUAL TABLE IF NOT EXISTS vec_table using vec0(
-                vector_value float[{config.online_store.vector_len}]
+                vector_value float[{vector_field_length}]
             );
             """
         )
@@ -476,18 +485,19 @@ def retrieve_online_documents_v2(
         conn = self._get_conn(config)
         cur = conn.cursor()
 
-        if online_store.vector_enabled and not online_store.vector_len:
-            raise ValueError("vector_len is not configured in the online store config")
+        vector_field_length = getattr(
+            _get_feature_view_vector_field_metadata(table), "vector_length", 512
+        )
 
         table_name = _table_id(config.project, table)
         vector_field = _get_vector_field(table)
 
         if online_store.vector_enabled:
-            query_embedding_bin = serialize_f32(query, online_store.vector_len)  # type: ignore
+            query_embedding_bin = serialize_f32(query, vector_field_length)  # type: ignore
             cur.execute(
                 f"""
                 CREATE VIRTUAL TABLE IF NOT EXISTS vec_table using vec0(
-                    vector_value float[{online_store.vector_len}]
+                    vector_value float[{vector_field_length}]
                 );
                 """
             )
Original file line number	Diff line number	Diff line change
`@@ -42,4 +42,7 @@ message FeatureSpecV2 {`
`42`	`42`
`43`	`43`	`// Metric used for vector similarity search.`
`44`	`44`	`string vector_search_metric = 6;`
	`45`	`+`
	`46`	`+ // Field indicating the vector length`
	`47`	`+ int32 vector_length = 7;`
`45`	`48`	`}`