core[patch], openai[patch]: Handle OpenAI developer msg (#28794)

- Convert developer openai messages to SystemMessage
- store additional_kwargs={"__openai_role__": "developer"} so that the
correct role can be reconstructed if needed
- update ChatOpenAI to read in openai_role

---------

Co-authored-by: Erick Friis <erick@langchain.dev>

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

@@ -139,8 +139,17 @@ def _convert_dict_to_message(_dict: Mapping[str, Any]) -> BaseMessage:
             tool_calls=tool_calls,
             invalid_tool_calls=invalid_tool_calls,
         )
-    elif role == "system":
-        return SystemMessage(content=_dict.get("content", ""), name=name, id=id_)
+    elif role in ("system", "developer"):
+        if role == "developer":
+            additional_kwargs = {"__openai_role__": role}
+        else:
+            additional_kwargs = {}
+        return SystemMessage(
+            content=_dict.get("content", ""),
+            name=name,
+            id=id_,
+            additional_kwargs=additional_kwargs,
+        )
     elif role == "function":
         return FunctionMessage(
             content=_dict.get("content", ""), name=cast(str, _dict.get("name")), id=id_
@@ -233,7 +242,9 @@ def _convert_message_to_dict(message: BaseMessage) -> dict:
             )
             message_dict["audio"] = audio
     elif isinstance(message, SystemMessage):
-        message_dict["role"] = "system"
+        message_dict["role"] = message.additional_kwargs.get(
+            "__openai_role__", "system"
+        )
     elif isinstance(message, FunctionMessage):
         message_dict["role"] = "function"
     elif isinstance(message, ToolMessage):
@@ -284,8 +295,14 @@ def _convert_delta_to_message_chunk(
             id=id_,
             tool_call_chunks=tool_call_chunks,  # type: ignore[arg-type]
         )
-    elif role == "system" or default_class == SystemMessageChunk:
-        return SystemMessageChunk(content=content, id=id_)
+    elif role in ("system", "developer") or default_class == SystemMessageChunk:
+        if role == "developer":
+            additional_kwargs = {"__openai_role__": "developer"}
+        else:
+            additional_kwargs = {}
+        return SystemMessageChunk(
+            content=content, id=id_, additional_kwargs=additional_kwargs
+        )
     elif role == "function" or default_class == FunctionMessageChunk:
         return FunctionMessageChunk(content=content, name=_dict["name"], id=id_)
     elif role == "tool" or default_class == ToolMessageChunk:

libs/partners/openai/tests/integration_tests/chat_models/test_base.py

@@ -1097,3 +1097,16 @@ def test_o1_max_tokens() -> None:
         "how are you"
     )
     assert isinstance(response, AIMessage)
+
+
+def test_developer_message() -> None:
+    llm = ChatOpenAI(model="o1", max_tokens=10)  # type: ignore[call-arg]
+    response = llm.invoke(
+        [
+            {"role": "developer", "content": "respond in all caps"},
+            {"role": "user", "content": "HOW ARE YOU"},
+        ]
+    )
+    assert isinstance(response, AIMessage)
+    assert isinstance(response.content, str)
+    assert response.content.upper() == response.content

libs/partners/openai/tests/unit_tests/chat_models/test_base.py

@@ -100,6 +100,16 @@ def test__convert_dict_to_message_system() -> None:
     assert _convert_message_to_dict(expected_output) == message
 
 
+def test__convert_dict_to_message_developer() -> None:
+    message = {"role": "developer", "content": "foo"}
+    result = _convert_dict_to_message(message)
+    expected_output = SystemMessage(
+        content="foo", additional_kwargs={"__openai_role__": "developer"}
+    )
+    assert result == expected_output
+    assert _convert_message_to_dict(expected_output) == message
+
+
 def test__convert_dict_to_message_system_with_name() -> None:
     message = {"role": "system", "content": "foo", "name": "test"}
     result = _convert_dict_to_message(message)
@@ -850,3 +860,25 @@ def test_nested_structured_output_strict() -> None:
         self_evaluation: SelfEvaluation
 
     llm.with_structured_output(JokeWithEvaluation, method="json_schema")
+
+
+def test__get_request_payload() -> None:
+    llm = ChatOpenAI(model="gpt-4o-2024-08-06")
+    messages: list = [
+        SystemMessage("hello"),
+        SystemMessage("bye", additional_kwargs={"__openai_role__": "developer"}),
+        {"role": "human", "content": "how are you"},
+    ]
+    expected = {
+        "messages": [
+            {"role": "system", "content": "hello"},
+            {"role": "developer", "content": "bye"},
+            {"role": "user", "content": "how are you"},
+        ],
+        "model": "gpt-4o-2024-08-06",
+        "stream": False,
+        "n": 1,
+        "temperature": 0.7,
+    }
+    payload = llm._get_request_payload(messages)
+    assert payload == expected