diff --git a/langchain_weaviate/vectorstores.py b/langchain_weaviate/vectorstores.py
index 77e2432..19eca88 100644
--- a/langchain_weaviate/vectorstores.py
+++ b/langchain_weaviate/vectorstores.py
@@ -157,7 +157,7 @@ def add_texts(
                 f"Tenant {tenant} does not exist in index {self._index_name}. "
                 "Creating tenant."
             )
-            tenant_objs = [weaviate.classes.Tenant(name=tenant)]
+            tenant_objs = [weaviate.classes.tenants.Tenant(name=tenant)]
             self._collection.tenants.create(tenants=tenant_objs)
 
         ids = []
@@ -167,7 +167,7 @@ def add_texts(
                 texts = list(texts)
             embeddings = self._embedding.embed_documents(texts)
 
-        with self._client.batch as batch:
+        with self._client.batch.dynamic() as batch:
             for i, text in enumerate(texts):
                 data_properties = {self._text_key: text}
                 if metadatas is not None:
@@ -259,7 +259,7 @@ def _perform_search(
             merged_props = {
                 **obj.properties,
                 **filtered_metadata,
-                **({"vector": obj.vector} if obj.vector else {}),
+                **({"vector": obj.vector["default"]} if obj.vector else {}),
             }
             doc = Document(page_content=text, metadata=merged_props)
             if not return_score:
@@ -486,7 +486,7 @@ def delete(
         if ids is None:
             raise ValueError("No ids provided to delete.")
 
-        id_filter = weaviate.classes.Filter.by_id().contains_any(ids)
+        id_filter = weaviate.classes.query.Filter.by_id().contains_any(ids)
 
         with self._tenant_context(tenant) as collection:
             collection.data.delete_many(where=id_filter)
diff --git a/tests/integration_tests/test_vectorstores.py b/tests/integration_tests/test_vectorstores.py
index a043c1d..471adaf 100644
--- a/tests/integration_tests/test_vectorstores.py
+++ b/tests/integration_tests/test_vectorstores.py
@@ -111,7 +111,7 @@ def test_similarity_search_with_metadata_and_filter(
         texts, embedding_openai, metadatas=metadatas, client=weaviate_client
     )
     output = docsearch.similarity_search(
-        "foo", k=2, filters=weaviate.classes.Filter.by_property("page").equal(0)
+        "foo", k=2, filters=weaviate.classes.query.Filter.by_property("page").equal(0)
     )
     assert output == [Document(page_content="foo", metadata={"page": 0})]
 
@@ -240,7 +240,7 @@ def test_max_marginal_relevance_search_with_filter(
         texts, embedding_openai, metadatas=metadatas, client=weaviate_client
     )
 
-    is_page_0_filter = weaviate.classes.Filter.by_property("page").equal(0)
+    is_page_0_filter = weaviate.classes.query.Filter.by_property("page").equal(0)
     # if lambda=1 the algorithm should be equivalent to standard ranking
     standard_ranking = docsearch.similarity_search("foo", k=2, filters=is_page_0_filter)
     output = docsearch.max_marginal_relevance_search(
@@ -444,7 +444,7 @@ def test_tenant_exists(
 ) -> None:
     index_name = "TestTenant"
     tenant_name = "Foo"
-    tenant = weaviate.classes.Tenant(name=tenant_name)
+    tenant = weaviate.classes.tenants.Tenant(name=tenant_name)
 
     # a collection with mt enabled
     docsearch_with_mt = WeaviateVectorStore(