Agno — Lightweight AI Agent Framework for Python

What is Agno?

Agno (formerly Phidata) is a lightweight Python framework for building AI agents with minimal boilerplate. It supports 23+ model providers, built-in tools (web search, file ops, SQL), knowledge bases, memory, and multi-agent teams — all in a clean, composable API.

Answer-Ready: Agno is a lightweight AI agent framework for Python (formerly Phidata) supporting 23+ model providers, built-in tools, knowledge bases, memory, and multi-agent teams. Build production agents in 5 lines of code. 20k+ GitHub stars.

Best for: Python developers who want a simple yet powerful agent framework. Works with: Claude, GPT, Gemini, Groq, Ollama, and 18+ more providers. Setup time: Under 1 minute.

Core Features

1. Multi-Provider Support

from agno.models.anthropic import Claude
from agno.models.openai import OpenAIChat
from agno.models.google import Gemini
from agno.models.ollama import Ollama

agent = Agent(model=Claude(id="claude-sonnet-4-20250514"))
agent = Agent(model=OpenAIChat(id="gpt-4o"))
agent = Agent(model=Gemini(id="gemini-2.5-pro"))
agent = Agent(model=Ollama(id="llama3"))

2. Built-In Tools

from agno.agent import Agent
from agno.tools.duckduckgo import DuckDuckGoTools
from agno.tools.yfinance import YFinanceTools

agent = Agent(
    model=Claude(id="claude-sonnet-4-20250514"),
    tools=[DuckDuckGoTools(), YFinanceTools()],
    show_tool_calls=True,
)
agent.print_response("What is AAPL stock price and latest news?")

Available tools: DuckDuckGo, YFinance, Newspaper, Shell, Python, SQL, File, Email, Slack, and more.

3. Knowledge Bases (RAG)

from agno.agent import Agent
from agno.knowledge.pdf import PDFKnowledgeBase
from agno.vectordb.pgvector import PgVector

knowledge = PDFKnowledgeBase(
    path="docs/",
    vector_db=PgVector(table_name="pdf_docs", db_url="postgresql://..."),
)

agent = Agent(knowledge=knowledge, search_knowledge=True)
agent.print_response("What does the contract say about termination?")

4. Memory & Storage

from agno.agent import Agent
from agno.memory.db.postgres import PgMemory

agent = Agent(
    model=Claude(id="claude-sonnet-4-20250514"),
    memory=PgMemory(db_url="postgresql://..."),
    enable_memory=True,
)
# Agent remembers past conversations

5. Multi-Agent Teams

from agno.agent import Agent
from agno.team import Team

researcher = Agent(name="Researcher", role="Research topics", tools=[DuckDuckGoTools()])
writer = Agent(name="Writer", role="Write articles based on research")

team = Team(
    agents=[researcher, writer],
    mode="coordinate",
)
team.print_response("Write an article about AI agents in 2026")

6. Structured Outputs

from pydantic import BaseModel

class MovieReview(BaseModel):
    title: str
    rating: float
    summary: str

agent = Agent(model=Claude(id="claude-sonnet-4-20250514"), response_model=MovieReview)
review = agent.run("Review the movie Inception")
print(review.rating)  # 9.2

FAQ

Q: What happened to Phidata? A: Phidata was renamed to Agno in early 2026. The API is largely the same with improvements.

Q: How does it compare to LangChain? A: Agno is more lightweight and opinionated. Less abstraction, faster to get started, but fewer integrations than LangChain.

Q: Is it production ready? A: Yes, used in production by many teams. The framework is mature (originally Phidata, 2+ years old).