From a5b86d3b2fe9383b600930d8de65c0d5a5b813f8 Mon Sep 17 00:00:00 2001
From: Cursor Agent <cursoragent@cursor.com>
Date: Fri, 13 Mar 2026 00:57:18 +0000
Subject: [PATCH] fix: revert realtime endpoint change, replace fragile asserts
 with fallback

- Revert realtime_endpoints/endpoints.py to original Response return
  (preserves backwards-compatible API contract; accepts 1 known mypy error)
- Replace 'assert provider_config is not None' with proper if/else
  fallback that re-raises the original exception when provider_config
  is None, avoiding AssertionError in production and python -O issues

Co-authored-by: yuneng-jiang <yuneng-jiang@users.noreply.github.com>
---
 litellm/llms/custom_httpx/llm_http_handler.py | 22 ++++++++++---------
 litellm/proxy/realtime_endpoints/endpoints.py |  5 +++--
 2 files changed, 15 insertions(+), 12 deletions(-)

diff --git a/litellm/llms/custom_httpx/llm_http_handler.py b/litellm/llms/custom_httpx/llm_http_handler.py
index 10caafb5ebe..a8d649064ac 100644
--- a/litellm/llms/custom_httpx/llm_http_handler.py
+++ b/litellm/llms/custom_httpx/llm_http_handler.py
@@ -4871,11 +4871,12 @@ class BaseLLMHTTPHandler:
                 timeout=timeout,
             )
         except Exception as e:
-            assert provider_config is not None
-            raise self._handle_error(
-                e=e,
-                provider_config=provider_config,
-            )
+            if provider_config is not None:
+                raise self._handle_error(
+                    e=e,
+                    provider_config=provider_config,
+                )
+            raise
 
     async def async_realtime_calls_handler(
         self,
@@ -4956,11 +4957,12 @@ class BaseLLMHTTPHandler:
                 timeout=timeout,
             )
         except Exception as e:
-            assert provider_config is not None
-            raise self._handle_error(
-                e=e,
-                provider_config=provider_config,
-            )
+            if provider_config is not None:
+                raise self._handle_error(
+                    e=e,
+                    provider_config=provider_config,
+                )
+            raise
 
     async def async_responses_websocket(
         self,
diff --git a/litellm/proxy/realtime_endpoints/endpoints.py b/litellm/proxy/realtime_endpoints/endpoints.py
index 876ff330cc5..0a91112298a 100644
--- a/litellm/proxy/realtime_endpoints/endpoints.py
+++ b/litellm/proxy/realtime_endpoints/endpoints.py
@@ -181,9 +181,10 @@ async def create_realtime_client_secret(
             upstream_resp.status_code,
             upstream_resp.text,
         )
-        raise HTTPException(
+        return Response(
+            content=upstream_resp.content,
             status_code=upstream_resp.status_code,
-            detail=upstream_resp.text,
+            media_type="application/json",
         )
 
     upstream_json: dict = upstream_resp.json()