Perplexity with Agno

Agno is an open-source Python framework for building agents, teams, and workflows. Connect it to Perplexity's Agent API through Agno's OpenAI Responses-compatible model class.

pip install agno openai

Set your Perplexity API key as an environment variable:

export PERPLEXITY_API_KEY="your_api_key_here"

Use OpenAIResponses with the Agent API base URL and any supported Agent API model. Pass a research preset through extra_body so answers are grounded in live web results:

import os

from agno.agent import Agent
from agno.models.openai import OpenAIResponses

agent = Agent(
    model=OpenAIResponses(
        id="openai/gpt-5.6-luna",
        base_url="https://api.perplexity.ai/v1",
        api_key=os.environ["PERPLEXITY_API_KEY"],
        extra_body={"preset": "medium"},
    ),
    markdown=True,
)

agent.print_response(
    "What are the latest breakthroughs in fusion energy this year?",
    stream=True,
)

Swap id for any Agent API model to route third-party models through Perplexity. The Agent API also provides research-backed presets and additional built-in tools.