All solutionsFor developers

Use it on two devices, then clone the code that runs it

Rooms, phone calls and AI voice agents from one SDK and one bill. Ship the prebuilt room UI today or keep the engine and draw every pixel yourself, on MediaSFU Cloud or your own servers.

50,000 free minutes that never expire. No card to start.

  • React, Vue, Angular, React Native, Flutter, iOS, Android
  • Prebuilt or headless
  • 99.99% uptime SLA
More to try

Start from an app that already works

What you get

The hard parts of realtime, handled

  • Prebuilt or headlessShip the complete room UI, swap in your own components, or keep only the engine.
  • Every platform you shipReact, Vue, Angular, React Native, Expo, Flutter, Swift, Kotlin and Unity.
  • Phone and agents on the same roomAdd a SIP number or an AI voice agent to a room through the same account and bill.
  • Cloud or your own serversRun on MediaSFU Cloud, or point the same SDK at MediaSFU Open on your own infrastructure.
  • Widgets for quick winsClick-to-call, agents and meetings in one script tag while the custom build catches up.
  • Bills you can checkEvery charged minute shows who sent, who received, the quality and the rate. Export it as CSV.
What it costs

Published per-minute rates

Usage-based, per stream, per person receiving it, per minute. No seats.

  • Audio in a room or call$0.0001 per minute$1 buys 10,000 audio minutes.
  • HD video$0.0015 per minuteSD video is $0.00075.
  • AI voice agent+$0.002 per minuteYour AI providers bill you directly, with no markup.
  • Recording+$0.002 per audio minuteSD video recording is $0.006. Off unless you turn it on.
A real workload1:1 video consultsTwo people in HD both ways, 30 minutes each500 consults × $0.09 = $45 a month
  • 50,000 free minutes to start, with no card. They never expire.
  • Volume discounts 20% off from 2.5M minutes a month · 40% off from 7.5M minutes a month, applied automatically.

Includes media transport, agent runtime and recording. Your AI providers, phone numbers and carrier minutes bill separately and directly.

See full pricing and the calculator
Build it your way

Your first room in a few minutes

  • SDKInstall and render a roomGet a free key, pass it to the prebuilt room, and open it on two devices.npm install mediasfu-reactjsRead the SDK guides
  • StartersWorking apps on GitHubCalls, classroom, auction and watch-together, each the source of a live room.Browse the starters
  • WidgetsOne script tagSix web components: click-to-call, AI agent, web agent, meeting portal and two dashboards.See the widgets

Works withReactVueAngularReact NativeExpoFlutterSwift / iOSKotlin / AndroidUnity

Questions

What teams ask first

What can I build with the MediaSFU SDKs?

Video rooms, webinars and broadcasts, phone and SIP calling, AI voice agents, and live translation, from the same SDK and account.

Can I start with a widget and move to the SDK later?

Yes. Many teams embed a widget first, then move to the SDK or headless mode when they need full control of the interface.

Is the media end-to-end encrypted?

Yes. Media travels end to end between participants unless you turn on a feature that processes it on our servers, such as recording, AI notes or translation.

Can I run it on my own servers?

Yes. MediaSFU Open runs on your own infrastructure, and the same SDK can point at it instead of MediaSFU Cloud.

Talk to a person

Questions go to the team, not a queue

Ask about your build, a migration or your bill. Every service runs under a 99.99% uptime SLA, live on the status page.

Your first 50,000 free minutes are on us

They never expire. Get a key, open a room from your own code, and add a phone number or an agent when you need one.