[Structured output type hints] Type returned after invoking llm.with_structured_output(PydanticClass, method="json_schema") is unknown instead of PydanticClass

[ai-learner-00]

Checked other resources

• I added a very descriptive title to this question.
• I searched the LangChain documentation with the integrated search.
• I used the GitHub search to find a similar question and didn't find it.

Commit to Help

• I commit to help with one of those options 👆

Example Code

from langchain_openai import AzureChatOpenAI
from pydantic import BaseModel, Field

from app.config.config import load_config


async def test_structured_output():
    print("\n\n")

    class Joke(BaseModel):
        setup: str = Field(description="The setup of the joke")
        punchline: str = Field(description="The punchline to the joke")

    config = await load_config()

    llm = AzureChatOpenAI(
        azure_deployment="gpt-4o",
        temperature=0,
        azure_endpoint=config.azure_openai_endpoint_us,
        api_key=config.azure_openai_api_key_us,
        api_version="2024-08-01-preview",
    )

    structured_llm = llm.with_structured_output(Joke, method="json_schema")
    output = structured_llm.invoke("What is the powerhouse of the cell?") # (variable) output: _DictOrPydantic[Unknown]

    print(output)

Description

I am using the method llm.with_structured_output to obtain an instance of a pydantic class. However, the type of the output is _DictOrPydantic[Unknown] instead of Joke.

I am wondering if there is an issue with the type hints:

_BM = TypeVar("_BM", bound=BaseModel)
_DictOrPydanticClass = Union[Dict[str, Any], Type[_BM], Type]
_DictOrPydantic = Union[Dict, _BM]

    def with_structured_output(
        self,
        schema: Optional[_DictOrPydanticClass] = None,
        *,
        method: Literal[
            "function_calling", "json_mode", "json_schema"
        ] = "function_calling",
        include_raw: bool = False,
        strict: Optional[bool] = None,
        **kwargs: Any,
    ) -> Runnable[LanguageModelInput, _DictOrPydantic]:

For example, if I try to access .setup, I get a warning Cannot access attribute "setup" for class "Dict[Unknown, Unknown]" Attribute "setup" is unknown.

System Info

langchain==0.3.9
langchain-community==0.3.8
langchain-core==0.3.21
langchain-openai==0.2.10
langchain-text-splitters==0.3.2

Windows WSL (dev container)
python 3.10

[DavidArmendariz]

Any updates on this? Having the same issue.

[marcusamp]

I'm wondering the same...