Agno
High-performance framework for building and running agents
Agno is profiled here as a LLM tool for engineering teams. Read about features, pricing, and how it compares to related options in the tools directory.
Description
Agno, formerly Phidata and created by Ashpreet Bedi, is an open-source Python framework for building multi-agent systems and running them in production. It centers on a lightweight Agent class with memory, knowledge, and tools, and pairs it with AgentOS, a FastAPI-based runtime that handles sessions, storage, and monitoring. Agno is built for fast agent creation and low memory overhead so a service can spin up many concurrent agents, and it stays model-agnostic across providers under the MPL-2.0 license.
Key Capabilities:
A lightweight Agent class with memory, knowledge, and tool use
AgentOS, a FastAPI runtime for sessions, storage, and monitoring
Teams and workflows for coordinating multiple agents on a task
Multimodal support across text, image, and audio inputs
Agentic retrieval that triggers knowledge lookups from agent reasoning
Model-agnostic integration across major LLM providers
Alternative tools
- Amazon Redshift
Cloud data warehouse for large-scale analytics on AWS
- smolagents
Lightweight agent library built around code-writing agents
- GraphRAG
Graph-based retrieval-augmented generation from Microsoft Research
- Reka
Multimodal models that reason across text, image, and video
- LLM Guard
Open-source security toolkit for LLM interactions
- Llama Guard
Open safeguard model for classifying LLM inputs and outputs