Perplexity search integration

Integrate with the Perplexity search tool using LangChain Python.

Perplexity Search is a web search API that returns ranked, source-attributed results designed for use by LLMs and agents. It powers the answer engine at perplexity.ai and is exposed through the dedicated Search API endpoint.

This page goes over how to use the Perplexity Search API as a LangChain tool.

Setup

Installation

Install the LangChain Perplexity integration package:

pip install -qU langchain-perplexity

# and some deps for this notebook
pip install -qU langchain langchain-openai

Credentials

You’ll need a Perplexity API key to use this integration. Create one in the Perplexity API key dashboard.

import getpass
import os

if not os.environ.get("PPLX_API_KEY"):
    os.environ["PPLX_API_KEY"] = getpass.getpass("Perplexity API key:\n")

The integration also accepts PERPLEXITY_API_KEY if you prefer that name.

Using PerplexitySearchResults tool

PerplexitySearchResults is a tool that can be used with LangChain agents to perform Perplexity searches. Calling it returns a JSON array of search results, each containing a title, url, snippet, date, and last_updated field.

from langchain_perplexity import PerplexitySearchResults

# Initialize the PerplexitySearchResults tool
search_tool = PerplexitySearchResults(max_results=5)

# Perform a search query
search_results = search_tool.invoke("When was the last time the New York Knicks won the NBA Championship?")

print("Search Results:", search_results)

Advanced features for PerplexitySearchResults

The tool accepts the following constructor and call-time arguments:

• max_results — number of results to return.
• country — ISO country code to bias results (e.g. "US").
• search_domain_filter — list of domains to include or exclude (max 20). Prefix a domain with - to exclude it. See the domain filter docs.
• search_recency_filter — one of "day", "week", "month", "year". See the date and time filter docs.
• search_after_date / search_before_date — date strings in MM/DD/YYYY format.

from langchain_perplexity import PerplexitySearchResults

# Restrict results to a recent timeframe and a specific set of domains
search_tool = PerplexitySearchResults(
    max_results=10,
    country="US",
    search_recency_filter="week",
    search_domain_filter=["nytimes.com", "reuters.com", "-pinterest.com"],
)

results = search_tool.invoke("Latest AI research papers")
print(results)

from langchain_perplexity import PerplexitySearchResults

# Restrict results to a specific date range
search_tool = PerplexitySearchResults(
    max_results=5,
    search_after_date="01/01/2025",
    search_before_date="06/30/2025",
)

results = search_tool.invoke("US Federal Reserve interest rate decisions")
print(results)

Use within an Agent

We can use the PerplexitySearchResults tool with a LangGraph agent. This gives the agent the ability to dynamically search the web for grounded, source-attributed information.

First, set up the language model. You’ll need to provide your OpenAI API key:

import getpass

if not os.environ.get("OPENAI_API_KEY"):
    os.environ["OPENAI_API_KEY"] = getpass.getpass("OpenAI API key:\n")

from langchain.agents import create_agent
from langchain.chat_models import init_chat_model
from langchain_perplexity import PerplexitySearchResults


# Initialize the language model
model = init_chat_model(model="gpt-5.5", temperature=0)

# Initialize the Perplexity search tool
perplexity_search = PerplexitySearchResults(
    max_results=5,
    search_recency_filter="month",
)

# Create the agent
agent = create_agent(model, [perplexity_search])

# Run the agent
user_input = "What are the latest developments in quantum computing?"

stream = agent.stream_events(
    {"messages": [{"role": "user", "content": user_input}]},
    version="v3",
)
for snapshot in stream.values:
    snapshot["messages"][-1].pretty_print()

---

API reference

For detailed documentation of the Perplexity Search API and all of its options, see the Search API reference and the Perplexity API documentation.

---

Connect these docs to Claude, VSCode, and more via MCP for real-time answers.

Edit this page on GitHub or file an issue.