v0.2.23 is now available

Lightweight WebRTC SFU
for Mobile, IoT, and Web.

Real-time application development shouldn't be complicated. PulseBeam is a general-purpose media server that connects clients without heavy architectures.

Built for Developers

Opinionated where it matters, flexible where you need it. PulseBeam is designed to be the simplest building block in your real-time stack.

Rust Powered Core

Memory-safe and concurrent by default. No garbage collector pauses, just predictable, low-latency performance.

HTTP Signaling

No WebSockets required. Use standard HTTP/REST for signaling (WHIP/WHEP compatible) with any backend.

Battery Efficient

Enforced H.264 (Baseline 4.1) and Opus ensures hardware acceleration on almost every mobile & IoT device.

Architecture for Scaling

Media flows through an optimized data plane, while the control plane stays separate so logic scales independently.

Optional SDKs

Client SDKs exist for convenience, not necessity. If your device speaks standard WebRTC, it works with PulseBeam.

Open Source

AGPL-3.0 for the server, Apache-2.0 for clients. Built in the open, hackable, and ready for your contributions.

Quickstart

The following quickstart assumes that you have a Linux machine or Docker installed. Get your first stream running in under a minute.

1

Run the PulseBeam Server

Use Docker to start the server instantly. No database or external dependencies required.

Open Ports

TCP/3000
HTTP Signaling
UDP/3478
WebRTC Traffic
bash
$docker run --rm --net=host ghcr.io/pulsebeamdev/pulsebeam:pulsebeam-v0.2.23
Download Binary|cargo run --release -p pulsebeam
2

Publish a Video

Publish your video through the live demo on the right.

Publisher
3

View the Stream

Connect a second peer to watch the broadcast. If you are running the server locally, you can use this viewer.

Once connected, you will see the video stream appear in the window on the right.

Viewer

Roadmap

Prototype

Working demo apps.

Media Engine

Simulcast & Congestion Control.

Core Features Current

Top-N audio selection, Data channel, Web Client SDK.

API & Integration

HTTP API & Webhooks (events).

Scaling

Multi-node / cascading SFU support.

Extensions

Recording, SIP, AI agents.