From bd6f2edbf30a70b7c9fefc6305f3d40a8a157d4f Mon Sep 17 00:00:00 2001
From: Krrish Dholakia <krrishdholakia@gmail.com>
Date: Sat, 16 Aug 2025 14:25:29 -0700
Subject: [PATCH] fix(streaming_handler.py): fix streaming chunk calculation

---
 litellm/litellm_core_utils/streaming_handler.py | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/litellm/litellm_core_utils/streaming_handler.py b/litellm/litellm_core_utils/streaming_handler.py
index 3721851a38f..968a6aade98 100644
--- a/litellm/litellm_core_utils/streaming_handler.py
+++ b/litellm/litellm_core_utils/streaming_handler.py
@@ -1731,6 +1731,11 @@ class CustomStreamWrapper:
                         # Create a new object without the removed attribute
                         processed_chunk = self.model_response_creator(chunk=obj_dict)
                     print_verbose(f"final returned processed chunk: {processed_chunk}")
+
+                    # add usage as hidden param
+                    if self.sent_last_chunk is True and self.stream_options is None:
+                        usage = calculate_total_usage(chunks=self.chunks)
+                        processed_chunk._hidden_params["usage"] = usage
                     return processed_chunk
                 raise StopAsyncIteration
             else:  # temporary patch for non-aiohttp async calls
@@ -1774,6 +1779,7 @@ class CustomStreamWrapper:
                     messages=self.messages,
                     logging_obj=self.logging_obj,
                 )
+
                 response = self.model_response_creator()
                 if complete_streaming_response is not None:
                     setattr(