PromptTemplate is a template system for building dynamic prompts in LangChain and other LLM-based tools.

It lets you define a prompt structure with placeholders, which can be filled with actual values at runtime.

📦 Example

from langchain.prompts import PromptTemplate

template = PromptTemplate(
    input_variables=["name", "age"],
    template="My name is {name} and I am {age} years old."
)

output = template.format(name="Abhi", age=30)
print(output)

🖨️ Output:

My name is Abhi and I am 30 years old.

🔧 Core Parameters

🛠️ Why Use PromptTemplate?

1. ✅ Reusable prompts with dynamic content

2. ✅ Separates prompt logic from data

3. ✅ Helps in prompt engineering with LLM chains

4. ✅ Works with output parsers, agents, and tools

🔍 Real Use Case (With Output Parser)

from langchain.prompts import PromptTemplate
from langchain.output_parsers import PydanticOutputParser
from pydantic import BaseModel

# Define model
class Person(BaseModel):
    name: str
    age: int

# Parser
parser = PydanticOutputParser(pydantic_object=Person)

# Prompt with format instructions
prompt = PromptTemplate(
    input_variables=["text"],
    template="""
    Extract the person's details from the text below.

    Text: {text}

    {format_instructions}
    """,
    partial_variables={"format_instructions": parser.get_format_instructions()}
)

formatted_prompt = prompt.format(text="My name is Abhi and I am 30.")
print(formatted_prompt)

🧩 Advanced Features

🪄 Partial Prompting

partial_prompt = prompt.partial(name="Abhi")
output = partial_prompt.format(age=30)

🛠 Custom Templates with Jinja2 (Advanced)

If you want more control (e.g., loops, conditions), LangChain supports Jinja2 templates too.

from langchain.prompts.prompt import PromptTemplate
from langchain.prompts.few_shot import FewShotPromptTemplate

🧪 PromptTemplate + Chain Example

from langchain.chat_models import ChatOpenAI
from langchain.chains import LLMChain

llm = ChatOpenAI()

prompt = PromptTemplate(
    input_variables=["product"],
    template="Write an advertisement for the following product: {product}"
)

chain = LLMChain(llm=llm, prompt=prompt)

response = chain.run("eco-friendly toothbrush")
print(response)

📌 Summary