qiankun · micro-frontend runtimeShip independently. Compose on demand.

Use loadMicroApp to mount front-end apps from different teams and frameworks wherever your product needs them. Each app keeps its own development and release cadence.

Load your first micro-app

See when qiankun fits

Live examples

A host application composing independently delivered micro-apps at runtime

Release without lockstep

Each micro-app keeps its stack, repository, and release cadence. The host composes them at runtime.

Control each instance

Mount into any HTMLElement with loadMicroApp, then update or unmount the returned instance when needed.

Coexist with fewer conflicts

JavaScript sandboxing, optional style isolation, and native ESM support help different stacks share one page.

Control one micro-app instance

Install qiankun in the main app:

bash

npm install qiankun@rc

v3 installs from the rc tag

qiankun 3.0 is a release candidate, so npm's latest tag still resolves to 2.x. Ask for @rc explicitly to get v3.

Mount after the container exists, then keep the returned handle for status and teardown:

import { loadMicroApp } from 'qiankun';

const container = document.getElementById('micro-app-slot');
if (!container) throw new Error('micro-app-slot not found');

const microApp = loadMicroApp({
  name: 'orders',
  entry: '//localhost:7101',
  container,
});

await microApp.mountPromise;

// When this part of the page is removed:
await microApp.unmount();

The micro-app exports bootstrap, mount, and unmount; qiankun loads it into the HTMLElement and drives those lifecycles. Follow Getting started for a complete runnable setup.

For applications whose lifetime should be driven entirely by the URL, see the route-based registerMicroApps and start alternative.

A live version of the repository's example apps — two hosts driving the same set of micro-apps — runs at examples.qiankunjs.com.