Skip to main content

​
System Requirements

  • Python: 3.7 or higher (3.12 recommended)
  • Operating System: Linux, macOS, or Windows
  • RAM: 2GB minimum, 8GB recommended for local LLMs

​
Quick Install

The simplest way to install Agno: 
pip install agno
This installs the core framework with minimal dependencies.

​
Install with Optional Dependencies

Agno uses optional dependencies to keep the base install lightweight. Install what you need:

​
For Production APIs (AgentOS)

pip install "agno[os]"
Includes: FastAPI, uvicorn, SQLAlchemy, PyJWT, OpenTelemetry, croniter

​
For Specific LLM Providers

pip install "agno[openai]"

​
For Databases

pip install "agno[postgres]"

​
For Vector Databases

pip install "agno[pgvector]"

​
For Tools and Integrations

pip install "agno[mcp]"

​
For Document Processing

pip install "agno[pdf]"

​
Combined Installation

Install multiple optional dependencies at once: 
pip install "agno[os,openai,postgres,pgvector,mcp]"

​
Install Everything

For development or testing with all integrations:
This installs 100+ packages and may have dependency conflicts. Only use for local testing.
pip install "agno[models,tools,storage,vectordbs,knowledge]"

​
Environment Setup

​
Set API Keys

Create a .env file in your project directory:
.env
# LLM Provider Keys
OPENAI_API_KEY=your-openai-key
ANTHROPIC_API_KEY=your-anthropic-key
GOOGLE_API_KEY=your-google-key

# Tool API Keys
TAVILY_API_KEY=your-tavily-key
EXA_API_KEY=your-exa-key

# Database URLs
DATABASE_URL=postgresql://user:password@localhost:5432/agno
VECTOR_DB_URL=postgresql://user:password@localhost:5432/agno

# Optional: Disable telemetry
AGNO_TELEMETRY=false
Agno automatically loads .env files using python-dotenv.

​
Verify Installation

Create a simple test file to verify your installation:
test_install.py
from agno.agent import Agent
from agno.models.openai import OpenAIChat

agent = Agent(
    name="Test Agent",
    model=OpenAIChat(id="gpt-4o"),
)

response = agent.run("Say hello")
print(response.content)
Run it: 
python test_install.py
If you see a response, you’re all set!

​
Database Setup

Agno provides a script to start PostgreSQL with PgVector using Docker: 
# Clone the Agno repository (or download the script)
git clone https://github.com/agno-agi/agno.git
cd agno

# Start PostgreSQL with PgVector
./cookbook/scripts/run_pgvector.sh
This starts a container with:
  • Host: localhost
  • Port: 5532
  • Database: ai
  • User: ai
  • Password: ai
If you have an existing PostgreSQL installation:
  1. Install the PgVector extension: 
    CREATE EXTENSION IF NOT EXISTS vector;
  2. Create a database: 
    CREATE DATABASE agno;
  3. Configure your connection: 
    from agno.db.postgres import PostgresDb

db = PostgresDb(
    db_url="postgresql+psycopg://user:password@localhost:5432/agno"
)

​
SQLite (Development Only)

For quick prototyping and development: 
from agno.db.sqlite import SqliteDb

db = SqliteDb(db_file="agno.db")
SQLite is single-threaded and not suitable for production. Use PostgreSQL for production deployments.

​
Using Virtual Environments

​
Using venv

# Create virtual environment
python -m venv .venv

# Activate (Linux/macOS)
source .venv/bin/activate

# Activate (Windows)
.venv\Scripts\activate

# Install Agno
pip install "agno[os,openai,postgres]"

​
Using uv (Fast Alternative)

# Install uv
curl -LsSf https://astral.sh/uv/install.sh | sh

# Create virtual environment and install
uv venv
source .venv/bin/activate
uv pip install "agno[os,openai,postgres]"

​
Using uvx (No Virtual Environment)

# Run commands directly without managing environments
uvx --with "agno[os]" --with openai fastapi dev app.py

​
IDE Integration

Add Agno docs as a context source in your IDE:

​
Cursor

Settings → Indexing & Docs → Add: 
https://docs.agno.com/llms-full.txt

​
VSCode / Windsurf

Same URL works in most AI-powered IDEs that support context indexing.

​
Upgrade Agno

To upgrade to the latest version: 
pip install --upgrade agno
To upgrade with optional dependencies: 
pip install --upgrade "agno[os,openai,postgres]"

​
Uninstall

pip uninstall agno

​
Common Installation Issues

If you encounter dependency conflicts:
  1. Create a fresh virtual environment
  2. Install only what you need
  3. Avoid installing agno[tests] or agno[demo] in production
python -m venv .venv-clean
source .venv-clean/bin/activate
pip install "agno[os,openai,postgres,pgvector]"
If PostgreSQL adapter installation fails, make sure you have build tools:Ubuntu/Debian:
sudo apt-get install libpq-dev python3-dev
macOS:
brew install postgresql
Windows: Install PostgreSQL from postgresql.org
Use --user flag to install in user directory:
pip install --user agno
Or use a virtual environment (recommended).
Make sure you’re in the correct virtual environment:
which python  # Should point to your venv
pip list | grep agno  # Verify agno is installed

​
Next Steps

Quickstart

Build your first agent in 5 minutes

Core Concepts

Learn the fundamental building blocks

Cookbook

Explore working examples and patterns

API Reference

Browse the complete API documentation

​
Telemetry

Agno logs which model providers are used to prioritize updates. No sensitive data is collected. To disable telemetry: 
export AGNO_TELEMETRY=false
Or in your .env file:
.env
AGNO_TELEMETRY=false

Build docs developers (and LLMs) love

Get started for freeTalk to us