diff --git a/pyproject.toml b/pyproject.toml
index 7b6efdc..551972f 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "turbopuffer"
-version = "0.1.23"
+version = "0.1.24"
 description = "Python Client for accessing the turbopuffer API"
 authors = ["turbopuffer Inc. <info@turbopuffer.com>"]
 homepage = "https://turbopuffer.com"
diff --git a/tests/test_backend.py b/tests/test_backend.py
index 3576659..64814aa 100644
--- a/tests/test_backend.py
+++ b/tests/test_backend.py
@@ -27,7 +27,7 @@ def test_500_retried():
 
     with mock.patch.object(time, 'sleep', return_value=None) as sleep:
         with pytest.raises(tpuf.error.APIError):
-            backend.make_api_request('vectors', payload={})
+            backend.make_api_request('namespaces', payload={})
         assert sleep.call_count == tpuf.max_retries - 1
 
 
@@ -38,5 +38,5 @@ def test_429_retried():
 
     with mock.patch.object(time, 'sleep', return_value=None) as sleep:
         with pytest.raises(tpuf.error.APIError):
-            backend.make_api_request('vectors', payload={})
+            backend.make_api_request('namespaces', payload={})
         assert sleep.call_count == tpuf.max_retries - 1
diff --git a/turbopuffer/namespace.py b/turbopuffer/namespace.py
index c72480b..7305ecc 100644
--- a/turbopuffer/namespace.py
+++ b/turbopuffer/namespace.py
@@ -112,7 +112,7 @@ def __eq__(self, other):
             return False
 
     def refresh_metadata(self):
-        response = self.backend.make_api_request('vectors', self.name, method='HEAD')
+        response = self.backend.make_api_request('namespaces', self.name, method='HEAD')
         status_code = response.get('status_code')
         if status_code == 200:
             headers = response.get('headers', dict())
@@ -170,7 +170,7 @@ def schema(self) -> NamespaceSchema:
         """
         Returns the current schema for the namespace.
         """
-        response = self.backend.make_api_request('vectors', self.name, 'schema', method='GET')
+        response = self.backend.make_api_request('namespaces', self.name, 'schema', method='GET')
         return parse_namespace_schema(response["content"])
 
     def update_schema(self, schema_updates: NamespaceSchema):
@@ -181,7 +181,7 @@ def update_schema(self, schema_updates: NamespaceSchema):
         See https://turbopuffer.com/docs/schema for specifics on allowed updates.
         """
         request_payload = json.dumps({key: value.as_dict() for key, value in schema_updates.items()}).encode()
-        response = self.backend.make_api_request('vectors', self.name, 'schema', method='POST', payload=request_payload)
+        response = self.backend.make_api_request('namespaces', self.name, 'schema', method='POST', payload=request_payload)
         return parse_namespace_schema(response["content"])
 
     @overload
@@ -255,7 +255,7 @@ def upsert(self, data=None, ids=None, vectors=None, attributes=None, schema=None
             if schema is not None:
                 payload["schema"] = schema
 
-            response = self.backend.make_api_request('vectors', self.name, payload=payload)
+            response = self.backend.make_api_request('namespaces', self.name, payload=payload)
 
             assert response.get('content', dict()).get('status', '') == 'OK', f'Invalid upsert() response: {response}'
             self.metadata = None  # Invalidate cached metadata
@@ -325,12 +325,12 @@ def delete(self, ids: Union[int, str, List[int], List[str]]) -> None:
         """
 
         if isinstance(ids, int) or isinstance(ids, str):
-            response = self.backend.make_api_request('vectors', self.name, payload={
+            response = self.backend.make_api_request('namespaces', self.name, payload={
                 'ids': [ids],
                 'vectors': [None],
             })
         elif isinstance(ids, list):
-            response = self.backend.make_api_request('vectors', self.name, payload={
+            response = self.backend.make_api_request('namespaces', self.name, payload={
                 'ids': ids,
                 'vectors': [None] * len(ids),
             })
@@ -391,7 +391,7 @@ def query(self,
             else:
                 raise ValueError(f'query() input type must be compatible with turbopuffer.VectorQuery: {type(query_data)}')
 
-        response = self.backend.make_api_request('vectors', self.name, 'query', payload=query_data.__dict__)
+        response = self.backend.make_api_request('namespaces', self.name, 'query', payload=query_data.__dict__)
         result = VectorResult(response.get('content', dict()), namespace=self)
         result.performance = response.get('performance')
         return result
@@ -404,7 +404,7 @@ def vectors(self, cursor: Optional[Cursor] = None) -> VectorResult:
         If you want to look up vectors by ID, use the query function with an id filter.
         """
 
-        response = self.backend.make_api_request('vectors', self.name, query={'cursor': cursor})
+        response = self.backend.make_api_request('namespaces', self.name, query={'cursor': cursor})
         content = response.get('content', dict())
         next_cursor = content.pop('next_cursor', None)
         result = VectorResult(content, namespace=self, next_cursor=next_cursor)
@@ -416,7 +416,7 @@ def delete_all_indexes(self) -> None:
         Deletes all indexes in a namespace.
         """
 
-        response = self.backend.make_api_request('vectors', self.name, 'index', method='DELETE')
+        response = self.backend.make_api_request('namespaces', self.name, 'index', method='DELETE')
         assert response.get('content', dict()).get('status', '') == 'ok', f'Invalid delete_all_indexes() response: {response}'
 
     def delete_all(self) -> None:
@@ -424,7 +424,7 @@ def delete_all(self) -> None:
         Deletes all data as well as all indexes.
         """
 
-        response = self.backend.make_api_request('vectors', self.name, method='DELETE')
+        response = self.backend.make_api_request('namespaces', self.name, method='DELETE')
         assert response.get('content', dict()).get('status', '') == 'ok', f'Invalid delete_all() response: {response}'
         self.metadata = None  # Invalidate cached metadata
 
@@ -438,7 +438,7 @@ def recall(self, num=20, top_k=10) -> float:
         Recall is calculated as the ratio of matching vectors between the two search results.
         """
 
-        response = self.backend.make_api_request('vectors', self.name, '_debug', 'recall', query={'num': num, 'top_k': top_k})
+        response = self.backend.make_api_request('namespaces', self.name, '_debug', 'recall', query={'num': num, 'top_k': top_k})
         content = response.get('content', dict())
         assert 'avg_recall' in content, f'Invalid recall() response: {response}'
         return float(content.get('avg_recall'))
@@ -525,7 +525,7 @@ def __next__(self):
             raise StopIteration
         else:
             response = self.backend.make_api_request(
-                'vectors',
+                'namespaces',
                 query={'cursor': self.next_cursor}
             )
             content = response.get('content', dict())
@@ -542,7 +542,7 @@ def namespaces(api_key: Optional[str] = None) -> Iterable[Namespace]:
     If no api_key is provided, the globally configured API key will be used.
     """
     backend = Backend(api_key)
-    response = backend.make_api_request('vectors')
+    response = backend.make_api_request('namespaces')
     content = response.get('content', dict())
     next_cursor = content.pop('next_cursor', None)
     return NamespaceIterator(backend, content.pop('namespaces', list()), next_cursor)
diff --git a/turbopuffer/version.py b/turbopuffer/version.py
index 6f1ee1b..df95409 100644
--- a/turbopuffer/version.py
+++ b/turbopuffer/version.py
@@ -1 +1 @@
-VERSION = '0.1.23'
+VERSION = '0.1.24'