What are output parsers in LangChain?

Output parsers sit at the end of a chain and transform the raw text or message returned by an LLM into a more structured or usable form. Without a parser, chain.invoke() returns an AIMessage object; with a parser, you get a plain string, a Python dict, a validated Pydantic model, or a list — whatever your downstream code expects.

The most common parsers:

• StrOutputParser — extracts .content from an AIMessage, returns a string. Used in virtually every chain: prompt | model | StrOutputParser()
• JsonOutputParser — parses the model's text as JSON and returns a Python dict. Works best when the prompt instructs the model to return valid JSON.
• PydanticOutputParser — validates parsed JSON against a Pydantic schema. The parser injects format instructions into the prompt automatically via parser.get_format_instructions().
• CommaSeparatedListOutputParser — splits a comma-delimited response into a Python list.
• StructuredOutputParser — uses a JSON schema for more flexible structured output.

from langchain_core.output_parsers import JsonOutputParser
from pydantic import BaseModel

class Person(BaseModel):
    name: str
    age: int

parser = JsonOutputParser(pydantic_object=Person)
chain = prompt | model | parser
result = chain.invoke({"query": "John is 30 years old"})
# result: {'name': 'John', 'age': 30}