From 697cb0906011cb84f2cc8c34b4ff7a2d7402dc13 Mon Sep 17 00:00:00 2001
From: AlexsanderHamir <alexsanderhamirgomesbaptista@gmail.com>
Date: Mon, 17 Nov 2025 10:51:18 -0800
Subject: [PATCH] fix: shared_sessions not being used

---
 litellm/llms/openai/openai.py | 5 +++++
 litellm/main.py               | 2 ++
 2 files changed, 7 insertions(+)

diff --git a/litellm/llms/openai/openai.py b/litellm/llms/openai/openai.py
index 2949e35e5e7..3282b7665c0 100644
--- a/litellm/llms/openai/openai.py
+++ b/litellm/llms/openai/openai.py
@@ -1414,6 +1414,7 @@ class OpenAIChatCompletion(BaseLLM, BaseOpenAILLM):
         timeout: Union[float, httpx.Timeout],
         aspeech: Optional[bool] = None,
         client=None,
+        shared_session: Optional["ClientSession"] = None,
     ) -> HttpxBinaryResponseContent:
         if aspeech is not None and aspeech is True:
             return self.async_audio_speech(
@@ -1428,6 +1429,7 @@ class OpenAIChatCompletion(BaseLLM, BaseOpenAILLM):
                 max_retries=max_retries,
                 timeout=timeout,
                 client=client,
+                shared_session=shared_session,
             )  # type: ignore
 
         openai_client = self._get_openai_client(
@@ -1437,6 +1439,7 @@ class OpenAIChatCompletion(BaseLLM, BaseOpenAILLM):
             timeout=timeout,
             max_retries=max_retries,
             client=client,
+            shared_session=shared_session,
         )
 
         response = cast(OpenAI, openai_client).audio.speech.create(
@@ -1460,6 +1463,7 @@ class OpenAIChatCompletion(BaseLLM, BaseOpenAILLM):
         max_retries: int,
         timeout: Union[float, httpx.Timeout],
         client=None,
+        shared_session: Optional["ClientSession"] = None,
     ) -> HttpxBinaryResponseContent:
         openai_client = cast(
             AsyncOpenAI,
@@ -1470,6 +1474,7 @@ class OpenAIChatCompletion(BaseLLM, BaseOpenAILLM):
                 timeout=timeout,
                 max_retries=max_retries,
                 client=client,
+                shared_session=shared_session,
             ),
         )
 
diff --git a/litellm/main.py b/litellm/main.py
index 14d0b04b7b5..412d7f1c38e 100644
--- a/litellm/main.py
+++ b/litellm/main.py
@@ -5747,6 +5747,7 @@ def speech(  # noqa: PLR0915
     proxy_server_request = kwargs.get("proxy_server_request", None)
     extra_headers = kwargs.get("extra_headers", None)
     model_info = kwargs.get("model_info", None)
+    shared_session = kwargs.get("shared_session", None)
     model, custom_llm_provider, dynamic_api_key, api_base = get_llm_provider(
         model=model, custom_llm_provider=custom_llm_provider, api_base=api_base
     )  # type: ignore
@@ -5856,6 +5857,7 @@ def speech(  # noqa: PLR0915
             timeout=timeout,
             client=client,  # pass AsyncOpenAI, OpenAI client
             aspeech=aspeech,
+            shared_session=shared_session,
         )
     elif custom_llm_provider == "azure":
         # Check if this is Azure Speech Service (Cognitive Services TTS)