Skip to main content

The runtime for agentic software

Build agents, teams, and workflows. Run them as scalable services. Monitor and manage them in production.

Get startedAPI reference

Quick start

Build and run your first agent in minutes

1

Install Agno

Install the package using pip or your preferred package manager.
pip install agno
2

Set your API key

Configure your model provider credentials as environment variables.
export ANTHROPIC_API_KEY="your-api-key"
3

Create your first agent

Build a simple agent with tools and run it.
from agno.agent import Agent
from agno.models.anthropic import Claude
from agno.tools.duckduckgo import DuckDuckGoTools

agent = Agent(
    model=Claude(id="claude-sonnet-4-5"),
    tools=[DuckDuckGoTools()]
)

agent.print_response("What's the latest news on AI?")
4

Run as a production API

Serve your agent as a scalable FastAPI service with AgentOS.
from agno.agent import Agent
from agno.os import AgentOS
from agno.models.anthropic import Claude

agent = Agent(
    name="My Agent",
    model=Claude(id="claude-sonnet-4-5")
)

agent_os = AgentOS(agents=[agent])
app = agent_os.get_app()
Run with uvicorn or your preferred ASGI server:
uvicorn app:app

Explore by topic

Learn about the core capabilities of Agno

Agents

Build autonomous agents with tools, memory, and knowledge

Teams

Coordinate multiple agents for complex tasks

Workflows

Create sequential and parallel execution pipelines

Knowledge & RAG

Add document retrieval and hybrid search

Memory & State

Manage user preferences and session state

AgentOS Runtime

Deploy agents as production APIs

Key features

Everything you need to build production-ready agentic systems

40+ model providers

OpenAI, Anthropic, Google, AWS Bedrock, Azure, and more. Switch providers with a single line of code.

100+ tool integrations

Web search, GitHub, databases, MCP servers, and custom tools. Give agents real-world capabilities.

Self-learning agents

Agents that improve over time through automated feedback loops and learning mechanisms.

Built for production

Stateless runtime, horizontal scaling, native tracing, and enterprise-grade security features.

Resources

Cookbook examples

Browse ready-to-run examples for common use cases

GitHub repository

Explore the source code and contribute

Ready to build?

Start building production-ready agentic systems in minutes

Get started now

Build docs developers (and LLMs) love

Get started for freeTalk to us