We don't publish
your competitive advantage.

IFUsing PydanticOutputParser with a pydantic v2 BaseModel subclass in LangChain >=0.1.6 raises ValidationError because the internal validator expects pydantic v1 BaseModel.

THENDowngrade LangChain to version 0.1.5 or earlier, or ensure your output schema inherits from pydantic v1 BaseModel (from pydantic.v1) instead of pydantic v2. For create_structured_output_runnable, pass a pydantic v1 model or use the downgraded version.

IFUsing PydanticOutputParser or create_structured_output_runnable with a pydantic v2 BaseModel in LangChain version 0.1.6 or later throws a ValidationError because the parser expects a pydantic v1 BaseModel.

THENTo resolve, use pydantic v1 BaseModel (e.g., from pydantic.v1) for the output schema. Alternatively, downgrade LangChain to 0.1.5 or use a different output parsing method like the newer structured output methods that support pydantic v2.

IFUsing PydanticOutputParser (or create_structured_output_runnable) with a pydantic v2 BaseModel in langchain >=0.1.6 causes ValidationError: 'subclass of BaseModel expected'.

THENDowngrade langchain to 0.1.5, or ensure your output model inherits from pydantic.v1.BaseModel (via langchain's pydantic v1 compatibility layer) instead of pydantic v2 BaseModel.

IFPydantic 2 migration broke Playwright tools like NavigateBackTool, CurrentWebPageTool, and ExtractTextTool causing KeyError: 'data' when creating an OpenAI tools agent.

THENAdd an explicit Pydantic BaseModel subclass (e.g., class NavigateBackInput(BaseModel): pass) and set args_schema to that class in the tool definition. Alternatively, downgrade langchain-core to <0.3 and related packages to versions before the breaking change (langchain 0.2.16, langchain-community 0.2.17, langchain-core 0.2.41, langchain-openai 0.1.25).

IFKeyError: 'data' when using Playwright's NavigateBackTool, CurrentWebPageTool, or ExtractTextTool with LangChain OpenAI tools agent due to Pydantic 2 mismatch.

THENDowngrade langchain, langchain-core, langchain-community, and langchain-openai to versions before the breaking change (e.g., langchain 0.2.16, langchain-community 0.2.17, langchain-core 0.2.41, langchain-openai 0.1.25). Alternatively, add an explicit Pydantic v2 BaseModel Input class to the tool's args_schema to override the empty schema.

IFUpgrading to langchain 0.1.6 or later when using PydanticOutputParser with a pydantic v2 BaseModel subclass causes a ValidationError because the parser now expects a pydantic v1 BaseModel subclass.

THENDowngrade to langchain 0.1.5, or define your Pydantic models using pydantic v1's BaseModel (e.g., from pydantic.v1 import BaseModel). Alternatively, use a different output parser that supports pydantic v2 or wrap your v2 models appropriately.

IFUsing with_fallbacks on a LangChain ChatOpenAI instance results in ValidationError: 'Can't instantiate abstract class BaseLanguageModel with abstract methods...'

THENEnsure consistent Pydantic version usage. Pin to Pydantic v1 (e.g., pip install pydantic==1.10.12) until LangChain's migration to v2 is complete. Do not mix Pydantic v1 and v2 code when using LangChain.