Documentation

How bēon works

Take your feed in over standard protocols, transcode it to an adaptive ladder, and deliver it sub-second from an edge we stand up in your region. Here's the whole path, honestly.

Your encoder—SRT/WHIP/RTMP→IngestABR transcodeRegional edge—WebRTC/LL-HLS→Your viewers

1 · Ingest

Push your feed with what you already use: SRT, WHIP or RTMP (including RTMPS) — that covers OBS, vMix, hardware encoders and browser production tools. NDI, SDI-over-IP and MPEG-TS are available on request. Ingest is free; you're never billed for what you send us, only for what we deliver.

2 · Transcode

We transcode to an adaptive bitrate (ABR) ladder so the feed stays watchable on campus Wi-Fi or a phone on a train, and deliver parallel streams (extra angles, commentary tracks) when your use case needs them.

3 · Delivery & latency

You pick the delivery mode against the trade-off you care about:

ModeGlass-to-glassBest for
WebRTC400–600 ms (p95 < 1s)betting, auctions, live casino, interactive shows
LL-HLS / CMAF2–4 sreach where browser compatibility matters most
Standard HLS15–30 slean-back, non-interactive

Our stack targets sub-500 ms glass-to-glass on WebRTC for the region you care about — measured, tunable and observable, not a brochure number.

4 · Access control & recording

Playback is gated with signed URLs per viewer, so only authorised people watch and you can revoke anytime. Turn on auto-VOD and the full session lands in your library for on-demand replay the moment the stream ends — the same material for live and replay audiences.

5 · Observability

A dashboard for the metrics customers actually ask for: concurrent viewers, regional breakdown, latency percentiles, ingest health and per-GB egress. Raw data is available via API so you can pipe it into your own BI.

6 · Where it runs — regions on demand

We don't pretend to run hundreds of always-on PoPs. We stand up a dedicated regional edge exactly where your traffic is — AWS, GCP, Azure, OVH, Hetzner, bare-metal, or your own cloud — usually within 48 hours. You pay for the capacity you use, not global over-provisioning.

Onboarding — white-glove

Today onboarding is hands-on by design, not a self-serve signup:

Test your own stream free → Book 20 minutes →

Questions: hello@beon.live · a real engineer replies, usually within one business hour.