`NotRequired` in TypedDict causes TypeError when converting to OpenAI function schema

[roombawulf]

Checked other resources

• This is a bug, not a usage question.
• I added a clear and descriptive title that summarizes this issue.
• I used the GitHub search to find a similar question and didn't find it.
• I am sure that this is a bug in LangChain rather than my code.
• The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package).
• This is not related to the langchain-community package.
• I posted a self-contained, minimal, reproducible example. A maintainer can copy it and run it AS IS.

Package (Required)

• langchain
• langchain-openai
• langchain-anthropic
• langchain-classic
• langchain-core
• langchain-cli
• langchain-model-profiles
• langchain-tests
• langchain-text-splitters
• langchain-chroma
• langchain-deepseek
• langchain-exa
• langchain-fireworks
• langchain-groq
• langchain-huggingface
• langchain-mistralai
• langchain-nomic
• langchain-ollama
• langchain-perplexity
• langchain-prompty
• langchain-qdrant
• langchain-xai
• Other / not sure / general

Example Code (Python)

from typing import NotRequired, TypedDict
from langchain_core.utils.function_calling import convert_to_openai_function

# Simple TypedDict with NotRequired field
class MyTypedDict(TypedDict):
    required_field: str
    optional_field: NotRequired[str]

# This should work but raises TypeError
result = convert_to_openai_function(MyTypedDict)

Error Message and Stack Trace (if applicable)

Traceback (most recent call last):
  File "<stdin>", line 8, in <module>
  File "/path/to/langchain_core/utils/function_calling.py", line 396, in convert_to_openai_function
    "dict", _convert_typed_dict_to_openai_function(cast("type", function))
  File "/path/to/langchain_core/utils/function_calling.py", line 213, in _convert_typed_dict_to_openai_function
    _convert_any_typed_dicts_to_pydantic(typed_dict, visited=visited),
  File "/path/to/langchain_core/utils/function_calling.py", line 261, in _convert_any_typed_dicts_to_pydantic
    new_arg_type = _convert_any_typed_dicts_to_pydantic(
  File "/path/to/langchain_core/utils/function_calling.py", line 278, in _convert_any_typed_dicts_to_pydantic
    return subscriptable_origin[type_args]  # type: ignore[index]
  File "/path/to/python3.12/typing.py", line 377, in inner
    return func(*args, **kwds)
  File "/path/to/python3.12/typing.py", line 489, in __getitem__
    return self._getitem(self, parameters)
  File "/path/to/python3.12/typing.py", line 2988, in NotRequired
    item = _type_check(parameters, f'{self._name} accepts only a single type.')
  File "/path/to/python3.12/typing.py", line 204, in _type_check
    raise TypeError(f"{msg} Got {arg!r:.100}.")
TypeError: NotRequired accepts only a single type. Got (<class 'str'>,).

Description

When using NotRequired fields in a TypedDict with LangChain's convert_to_openai_function() or .with_structured_output(), the conversion fails with a TypeError.

Caused by langchain_core/utils/function_calling.py (Ln 278)

return subscriptable_origin[type_args]

System Info

System Information

OS: macOS
Python Version: 3.12.0

Package Information

langchain_core: 1.0.7
langchain: 0.3.22
langsmith: 0.3.45
pydantic: 2.10.6
typing-extensions: 4.12.2