The .get_format_instructions() method returns a string prompt that tells an LLM how to format its output so that it can be parsed cleanly into a Pydantic model.

✅ Why It's Useful

When using LLMs with a structured output parser like PydanticOutputParser, you want the model to return a JSON that can be directly validated. .get_format_instructions() helps you tell the model exactly how to format that output.

🧪 Example

from langchain.output_parsers import PydanticOutputParser
from pydantic import BaseModel

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

parser = PydanticOutputParser(pydantic_object=Person)

print(parser.get_format_instructions())

🖨️ Output:

The output should be a JSON object with the following format:
{
  "name": string,
  "age": integer
}

📌 Use Case in Prompting

format_instructions = parser.get_format_instructions()

prompt = f"""
Extract the person's details from the text below.

Text: "My name is Abhi and I am 30 years old."

{format_instructions}
"""

So the full prompt sent to the LLM looks like:

Extract the person's details from the text below.

Text: "My name is Abhi and I am 30 years old."

The output should be a JSON object with the following format:
{
  "name": string,
  "age": integer
}

💡 Now the LLM is guided to return properly structured JSON!

✅ Summary