Skip to main content
NexAU is a general-purpose Python agent framework for building intelligent agents. Define your agent’s tools and LLM in a YAML file or in Python, then run it — locally, as an HTTP server, or as a CLI chat interface.

Quick Start

Build and run your first agent in minutes.

Installation

Install NexAU via pip, uv, or from source.

Agents

Learn how to create and configure agents.

Tools

Equip your agent with built-in and custom tools.

Multi-Agent

Coordinate teams of agents for complex tasks.

Transports

Expose your agent as an HTTP API or CLI chat.

​
What you can build

Research Agents

Agents that search the web, read pages, and synthesize information.

Code Agents

Agents that read, write, and execute code in a secure sandbox.

Automation Pipelines

Multi-agent teams that parallelize and coordinate complex workflows.

​
How it works

1

Install NexAU

Install the package from the GitHub release or from source using pip or uv.
2

Configure your LLM

Point NexAU at any OpenAI-compatible provider — OpenAI, Anthropic, Ollama, or your own endpoint.
3

Define tools and your agent

Use built-in tools or write your own. Configure the agent in YAML or Python.
4

Run

Run your agent from the CLI, call it directly in Python, or serve it as an HTTP API.