openai[patch]: fix async http client (#19164)

Fix #19116

libs/partners/openai/langchain_openai/chat_models/azure.py

@@ -185,12 +185,17 @@ class AzureChatOpenAI(ChatOpenAI):
             "max_retries": values["max_retries"],
             "default_headers": values["default_headers"],
             "default_query": values["default_query"],
-            "http_client": values["http_client"],
         }
-        values["client"] = openai.AzureOpenAI(**client_params).chat.completions
+        if not values.get("client"):
-        values["async_client"] = openai.AsyncAzureOpenAI(
+            sync_specific = {"http_client": values["http_client"]}
-            **client_params
+            values["client"] = openai.AzureOpenAI(
-        ).chat.completions
+                **client_params, **sync_specific
+            ).chat.completions
+        if not values.get("async_client"):
+            async_specific = {"http_client": values["http_async_client"]}
+            values["async_client"] = openai.AsyncAzureOpenAI(
+                **client_params, **async_specific
+            ).chat.completions
         return values
 
     @property

libs/partners/openai/langchain_openai/chat_models/base.py

@@ -313,7 +313,12 @@ class ChatOpenAI(BaseChatModel):
     # Configure a custom httpx client. See the
     # [httpx documentation](https://www.python-httpx.org/api/#client) for more details.
     http_client: Union[Any, None] = None
-    """Optional httpx.Client."""
+    """Optional httpx.Client. Only used for sync invocations. Must specify 
+        http_async_client as well if you'd like a custom client for async invocations.
+    """
+    http_async_client: Union[Any, None] = None
+    """Optional httpx.AsyncClient. Only used for async invocations. Must specify 
+        http_client as well if you'd like a custom client for sync invocations."""
 
     class Config:
         """Configuration for this pydantic object."""
@@ -369,14 +374,17 @@ class ChatOpenAI(BaseChatModel):
             "max_retries": values["max_retries"],
             "default_headers": values["default_headers"],
             "default_query": values["default_query"],
-            "http_client": values["http_client"],
         }
 
         if not values.get("client"):
-            values["client"] = openai.OpenAI(**client_params).chat.completions
+            sync_specific = {"http_client": values["http_client"]}
+            values["client"] = openai.OpenAI(
+                **client_params, **sync_specific
+            ).chat.completions
         if not values.get("async_client"):
+            async_specific = {"http_client": values["http_async_client"]}
             values["async_client"] = openai.AsyncOpenAI(
-                **client_params
+                **client_params, **async_specific
             ).chat.completions
         return values
 

libs/partners/openai/langchain_openai/embeddings/azure.py

@@ -139,10 +139,17 @@ class AzureOpenAIEmbeddings(OpenAIEmbeddings):
             "max_retries": values["max_retries"],
             "default_headers": values["default_headers"],
             "default_query": values["default_query"],
-            "http_client": values["http_client"],
         }
-        values["client"] = openai.AzureOpenAI(**client_params).embeddings
+        if not values.get("client"):
-        values["async_client"] = openai.AsyncAzureOpenAI(**client_params).embeddings
+            sync_specific = {"http_client": values["http_client"]}
+            values["client"] = openai.AzureOpenAI(
+                **client_params, **sync_specific
+            ).embeddings
+        if not values.get("async_client"):
+            async_specific = {"http_client": values["http_async_client"]}
+            values["async_client"] = openai.AsyncAzureOpenAI(
+                **client_params, **async_specific
+            ).embeddings
         return values
 
     @property

libs/partners/openai/langchain_openai/embeddings/base.py

@@ -123,7 +123,12 @@ class OpenAIEmbeddings(BaseModel, Embeddings):
     retry_max_seconds: int = 20
     """Max number of seconds to wait between retries"""
     http_client: Union[Any, None] = None
-    """Optional httpx.Client."""
+    """Optional httpx.Client. Only used for sync invocations. Must specify 
+        http_async_client as well if you'd like a custom client for async invocations.
+    """
+    http_async_client: Union[Any, None] = None
+    """Optional httpx.AsyncClient. Only used for async invocations. Must specify 
+        http_client as well if you'd like a custom client for sync invocations."""
 
     class Config:
         """Configuration for this pydantic object."""
@@ -218,12 +223,17 @@ class OpenAIEmbeddings(BaseModel, Embeddings):
             "max_retries": values["max_retries"],
             "default_headers": values["default_headers"],
             "default_query": values["default_query"],
-            "http_client": values["http_client"],
         }
         if not values.get("client"):
-            values["client"] = openai.OpenAI(**client_params).embeddings
+            sync_specific = {"http_client": values["http_client"]}
+            values["client"] = openai.OpenAI(
+                **client_params, **sync_specific
+            ).embeddings
         if not values.get("async_client"):
-            values["async_client"] = openai.AsyncOpenAI(**client_params).embeddings
+            async_specific = {"http_client": values["http_async_client"]}
+            values["async_client"] = openai.AsyncOpenAI(
+                **client_params, **async_specific
+            ).embeddings
         return values
 
     @property

libs/partners/openai/langchain_openai/llms/azure.py

@@ -160,10 +160,17 @@ class AzureOpenAI(BaseOpenAI):
             "max_retries": values["max_retries"],
             "default_headers": values["default_headers"],
             "default_query": values["default_query"],
-            "http_client": values["http_client"],
         }
-        values["client"] = openai.AzureOpenAI(**client_params).completions
+        if not values.get("client"):
-        values["async_client"] = openai.AsyncAzureOpenAI(**client_params).completions
+            sync_specific = {"http_client": values["http_client"]}
+            values["client"] = openai.AzureOpenAI(
+                **client_params, **sync_specific
+            ).completions
+        if not values.get("async_client"):
+            async_specific = {"http_client": values["http_async_client"]}
+            values["async_client"] = openai.AsyncAzureOpenAI(
+                **client_params, **async_specific
+            ).completions
 
         return values
 

libs/partners/openai/langchain_openai/llms/base.py

@@ -149,7 +149,12 @@ class BaseOpenAI(BaseLLM):
     # Configure a custom httpx client. See the
     # [httpx documentation](https://www.python-httpx.org/api/#client) for more details.
     http_client: Union[Any, None] = None
-    """Optional httpx.Client."""
+    """Optional httpx.Client. Only used for sync invocations. Must specify 
+        http_async_client as well if you'd like a custom client for async invocations.
+    """
+    http_async_client: Union[Any, None] = None
+    """Optional httpx.AsyncClient. Only used for async invocations. Must specify 
+        http_client as well if you'd like a custom client for sync invocations."""
 
     class Config:
         """Configuration for this pydantic object."""
@@ -209,12 +214,17 @@ class BaseOpenAI(BaseLLM):
             "max_retries": values["max_retries"],
             "default_headers": values["default_headers"],
             "default_query": values["default_query"],
-            "http_client": values["http_client"],
         }
         if not values.get("client"):
-            values["client"] = openai.OpenAI(**client_params).completions
+            sync_specific = {"http_client": values["http_client"]}
+            values["client"] = openai.OpenAI(
+                **client_params, **sync_specific
+            ).completions
         if not values.get("async_client"):
-            values["async_client"] = openai.AsyncOpenAI(**client_params).completions
+            async_specific = {"http_client": values["http_async_client"]}
+            values["async_client"] = openai.AsyncOpenAI(
+                **client_params, **async_specific
+            ).completions
 
         return values