What's the best way to use """with_structured_output""" on Anthropic model???

[sangwooklee0201]

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 typing import Optional

from pydantic import BaseModel, Field


# Pydantic
class Joke(BaseModel):
    """Joke to tell user."""

    setup: str = Field(description="The setup of the joke")
    punchline: str = Field(description="The punchline to the joke")
    rating: Optional[int] = Field(
        default=None, description="How funny the joke is, from 1 to 10"
    )

sonnet_3_7_model = ChatAnthropic(
    anthropic_api_url=...,
    api_key=...,
    model_name=...,
    max_tokens=20000,
    thinking={"type": "enabled", "budget_tokens": 10000},
)    
joke_model = sonnet_3_7_model.with_structured_output(Joke)

joke_model.invoke("Tell me a joke about cats")

Description

I am a noob, so am following some langchain cookbook especially this one:

from typing import Optional

from pydantic import BaseModel, Field

gpt_4o_model = AzureChatOpenAI(
    api_key=...,
    azure_endpoint=...,
    openai_api_version=...,
    azure_deployment=GPT_4O_MODEL_NAME,
)
class Joke(BaseModel):
    """Joke to tell user."""

    setup: str = Field(description="The setup of the joke")
    punchline: str = Field(description="The punchline to the joke")
    rating: Optional[int] = Field(
        default=None, description="How funny the joke is, from 1 to 10"
    )
    
joking_model = gpt_4o_model.with_structured_output(Joke)

joking_model.invoke("Tell me a joke about cats")

When I ran it, it ran fine, output looks good, following Joke model.

But when I try it with "ChatAnthropic", I got following error:

sonnet_3_7_model = ChatAnthropic(
    anthropic_api_url=...,
    api_key=...,
    model_name=...,
    max_tokens=20000,
    thinking={"type": "enabled", "budget_tokens": 10000},
)

joking_model = sonnet_3_7_model.with_structured_output(Joke)

joking_model.invoke("Tell me a joke about cats")

UnprocessableEntityError: Error code: 422 - {'detail': [{'type': 'missing', 'loc': ['body', 'tools', 0, 'input_schema', 'properties', 'rating', 'type'], 'msg': 'Field required', 'input': {'anyOf': [{'type': 'integer'}, {'type': 'null'}], 'default': None, 'description': 'How funny the joke is, from 1 to 10'}}]}

What am I doing wrong? What should I do to get the correct answer like I did with Open AI model?

System Info

langchain==0.3.11
langchain-anthropic==0.3.10
langchain-chroma==0.2.0
langchain-community==0.3.11
langchain-core==0.3.49
langchain-openai==0.2.12
langchain-text-splitters==0.3.4