From 9b3190d07bc1cd65d6b01560548c49a738985ca4 Mon Sep 17 00:00:00 2001
From: mateo-berri <277851410+mateo-berri@users.noreply.github.com>
Date: Mon, 2 Nov 2026 16:05:01 -0800
Subject: [PATCH] fix(responses): keep a background poll's first terminal
 status and take a first-chunk error code from its HTTP status

---
 .../response_polling/background_streaming.py  | 10 +++----
 .../test_response_polling_handler.py          | 27 +++++++++++++++++++
 2 files changed, 30 insertions(+), 7 deletions(-)

diff --git a/litellm/proxy/response_polling/background_streaming.py b/litellm/proxy/response_polling/background_streaming.py
index 73190ad8df8..76f713ae963 100644
--- a/litellm/proxy/response_polling/background_streaming.py
+++ b/litellm/proxy/response_polling/background_streaming.py
@@ -111,7 +111,8 @@ def _first_chunk_error(response: object) -> Mapping[str, object] | None:
         return None
     if response.status_code < HTTPStatus.BAD_REQUEST:
         return None
-    return responses_error_fields(sse_error_object(bytes(response.body)) or {"code": response.status_code})
+    error_obj: Final = sse_error_object(bytes(response.body)) or {}
+    return responses_error_fields({**error_obj, "code": error_obj.get("code") or str(response.status_code)})
 
 
 async def _never_receive() -> Message:
@@ -387,12 +388,7 @@ async def background_streaming_task(
                                 status="in_progress",
                             )
 
-                        elif event_type in (
-                            "response.completed",
-                            "response.failed",
-                            "response.incomplete",
-                            "response.cancelled",
-                        ):
+                        elif terminal_status is None and event_type in _event_to_status:
                             # Terminal event - extract all ResponsesAPIResponse fields
                             # https://platform.openai.com/docs/api-reference/responses-streaming
                             finished_at = perf_counter()
diff --git a/tests/unit/proxy/response_polling/test_response_polling_handler.py b/tests/unit/proxy/response_polling/test_response_polling_handler.py
index 752dfc5f124..01fb8eac842 100644
--- a/tests/unit/proxy/response_polling/test_response_polling_handler.py
+++ b/tests/unit/proxy/response_polling/test_response_polling_handler.py
@@ -1736,6 +1736,26 @@ class TestBackgroundStreamingTerminalEvents:
                 None,
                 None,
             ),
+            (
+                (
+                    'data: {"type": "response.completed", "response": {"status": "completed", "usage": {"input_tokens": 3}}}\n\n',
+                    'data: {"type": "response.failed", "response": {"status": "failed", '
+                    '"error": {"code": "server_error", "message": "late"}}}\n\n',
+                ),
+                "completed",
+                None,
+                {"input_tokens": 3},
+            ),
+            (
+                (
+                    'data: {"type": "response.failed", "response": {"status": "failed", '
+                    '"error": {"code": "server_error", "message": "first"}, "usage": {"input_tokens": 3}}}\n\n',
+                    'data: {"type": "response.completed", "response": {"status": "completed", "usage": {"input_tokens": 9}}}\n\n',
+                ),
+                "failed",
+                {"code": "server_error", "message": "first"},
+                {"input_tokens": 3},
+            ),
         ],
         ids=[
             "error_frame_after_a_delta",
@@ -1753,6 +1773,8 @@ class TestBackgroundStreamingTerminalEvents:
             "done_behind_a_delta_in_one_chunk_then_completed",
             "completed_then_failed_in_one_chunk",
             "event_line_sent_as_a_data_line_of_one_event",
+            "completed_then_failed_in_the_next_chunk",
+            "failed_then_completed_in_the_next_chunk",
         ],
     )
     async def test_a_poll_stores_what_its_stream_ended_with(
@@ -1837,6 +1859,11 @@ class TestBackgroundStreamingTerminalEvents:
                 json.dumps({"error": {"type": "guardrail_scan_error", "code": 403}}),
                 {"code": "permission_error", "message": "Forbidden"},
             ),
+            (
+                404,
+                json.dumps({"error": {"message": "model not found", "type": "invalid_request_error", "param": None}}),
+                {"code": "not_found_error", "message": "model not found"},
+            ),
         ],
     )
     async def test_a_first_chunk_error_polls_failed_like_a_late_one(