---
title: NoLag with AutoGen
description: "Use NoLag as the realtime coordination layer under AutoGen. Give multi-agent conversations shared state, task handoff, live observation, and human approval gates."
---

# NoLag with AutoGen

AutoGen is a framework for building applications with multiple agents that converse to solve a task. It is good at modelling agent-to-agent conversations and tool use within an application. NoLag is complementary: it is the realtime layer that connects those agents to each other, to your services, to a UI, and to a human, across processes and machines.

AutoGen structures the conversation; NoLag moves the messages between agents, your system, and the people watching.

## What each layer does

| | AutoGen | NoLag |
| --- | --- | --- |
| Role | Multi-agent conversation and tool use | Realtime transport and coordination |
| Scope | Within an application run | Across agents, services, humans, and UIs |
| Gives you | Conversable agents, group chat, tools | Dispatch, shared state, live observation, approval gates, presence |

## Why add NoLag under AutoGen

- **Distribute agents.** Run agents in separate processes or services and route messages between them over topics.
- **Share state in realtime.** Publish intermediate results so other agents, dashboards, or humans see them live.
- **Stream the conversation to a UI.** Show the exchange and decisions as they happen.
- **Human approval gates.** Pause a sensitive action until a person approves it.

## How it fits

Use the NoLag SDK from inside your AutoGen agents. Where an agent emits a message or needs input from elsewhere in your system, publish or subscribe on a NoLag topic. AutoGen keeps owning the conversation; NoLag carries the realtime communication.

## Get started

Read the [AI agents guide](/docs/agents), then the [5-minute quick start](/docs/getting-started). SDKs for [Python](/docs/sdks/python), [JavaScript](/docs/sdks/javascript), and [Go](/docs/sdks/go).