diff --git a/vespa/repository/api/ingestion/streamer.py b/vespa/repository/api/ingestion/streamer.py
index a094be36..64680b69 100644
--- a/vespa/repository/api/ingestion/streamer.py
+++ b/vespa/repository/api/ingestion/streamer.py
@@ -7,10 +7,10 @@
 import logging
 import os
 import urllib.parse
-import xml.etree.ElementTree
 import zipfile
 
 import ijson
+import lxml
 import requests
 from lxml.etree import iterparse
 
@@ -185,22 +185,20 @@ async def _parse_xml_stream(self, stream):
         """
         Asynchronous parser for XML streams.
         """
-        # Ensure the stream is read synchronously
-        with io.TextIOWrapper(stream, encoding="utf-8", errors="replace") as wrapper:
-            for event, elem in iterparse(wrapper, events=("end",)):
-                if elem.tag == "node":
-                    # Create a dictionary from element attributes
-                    elem_data = dict(elem.attrib)
-
-                    # Add child tag attributes (k, v) to the dictionary
-                    for tag in elem.findall("tag"):
-                        key = tag.attrib.get("k")
-                        value = tag.attrib.get("v")
-                        if key and value:
-                            elem_data[key] = value
-
-                    yield elem_data
-                    elem.clear()  # Free memory
+        for event, elem in lxml.etree.iterparse(stream, events=("end",)):
+            if elem.tag == "node":
+                # Create a dictionary from element attributes
+                elem_data = dict(elem.attrib)
+
+                # Add child tag attributes (k, v) to the dictionary
+                for tag in elem.findall("tag"):
+                    key = tag.attrib.get("k")
+                    value = tag.attrib.get("v")
+                    if key and value:
+                        elem_data[key] = value
+
+                yield elem_data
+                elem.clear()  # Free memory
 
     def _split_triple(self, line):
         parts = line.rstrip(' .').split(' ', 2)