diff --git a/client/clip_client/client.py b/client/clip_client/client.py
index 5176227ab..262246f82 100644
--- a/client/clip_client/client.py
+++ b/client/clip_client/client.py
@@ -188,7 +188,6 @@ def _iter_doc(
                     ),
                 )
 
-            d.tags['__ordered_by_CAS__'] = i
             results.append(d)
             yield d
 
@@ -212,10 +211,7 @@ def _gather_encode_result(self, response, results: 'DocumentArray'):
         from rich import filesize
 
         r = response.data.docs
-        for d in r:
-            index = int(d.tags['__ordered_by_CAS__'])
-            results[index].embedding = d.embedding
-            results[index].tags.pop('__ordered_by_CAS__')
+        results[r[:, 'id']].embeddings = r.embeddings
 
         if not self._pbar._tasks[self._r_task].started:
             self._pbar.start_task(self._r_task)
@@ -345,10 +341,7 @@ async def aencode(self, content, **kwargs):
             async for da in self._async_client.post(
                 **self._get_post_payload(content, results, kwargs)
             ):
-                for d in da:
-                    index = int(d.tags['__ordered_by_CAS__'])
-                    results[index].embedding = d.embedding
-                    results[index].tags.pop('__ordered_by_CAS__')
+                results[da[:, 'id']].embeddings = da.embeddings
 
                 if not self._pbar._tasks[self._r_task].started:
                     self._pbar.start_task(self._r_task)
@@ -395,7 +388,6 @@ def _iter_rank_docs(
                     ),
                 )
 
-            d.tags['__ordered_by_CAS__'] = i
             results.append(d)
             yield d
 
@@ -420,10 +412,7 @@ def _gather_rank_result(self, response, results: 'DocumentArray'):
         from rich import filesize
 
         r = response.data.docs
-        for d in r:
-            index = int(d.tags['__ordered_by_CAS__'])
-            results[index].matches = d.matches
-            results[index].tags.pop('__ordered_by_CAS__')
+        results[r[:, 'id']][:, 'matches'] = r[:, 'matches']
 
         if not self._pbar._tasks[self._r_task].started:
             self._pbar.start_task(self._r_task)
@@ -513,10 +502,7 @@ async def arank(self, docs: Iterable['Document'], **kwargs) -> 'DocumentArray':
             async for da in self._async_client.post(
                 **self._get_rank_payload(docs, results, kwargs)
             ):
-                for d in da:
-                    index = int(d.tags['__ordered_by_CAS__'])
-                    results[index].matches = d.matches
-                    results[index].tags.pop('__ordered_by_CAS__')
+                results[da[:, 'id']][:, 'matches'] = da[:, 'matches']
 
                 if not self._pbar._tasks[self._r_task].started:
                     self._pbar.start_task(self._r_task)