No corporate servers. No blockchains. Epheral relays.

Relays + Messages that Evaporate by Design

Disappearing messages is the trend, we just dunked on that with disappearing servers
LUM is a radically different architecture for private communication.
Where other apps log, store, and index, LUM platform simply forgets.
Each relay is a microservice that lives in memory, runs for about a week, then deletes itself and respawns.
Each relay can run (and rotate) on every cloud service, or personal/private servers.
You know how you login to your favorite messaging app, and all your conversations reappear? Your messages are on private servers indefinately, not with LUM.
Do you want AI to consume your chats? NO, then use LUM messaging platform.

Noteworthy

Own Your Communications: By design the entire system can be owned with little effort
Ephemeral First: All data stored in RAM, automatically purged on shutdown.
Local First: Hyper local messaging is possible, the messages never leave your building!
Local or Cloud: One binary relay deployable anywhere — K3s, Docker, or bare metal.
Configurable Encryption: Choose the cipher‑chain and number of encryption rounds per chat group.
Post‑Quantum Installed: All latest Post-Quantum Ciphers (and experimental) available to you
No Metadata: Relays never learn who you are — only which anonymous channel UUID to serve.

How It Works

Create a Channel: Each chat group forms its own isolated identity (UUID + cipher chain).
Deploy a Relay: Lightweight lum-relay container runs in memory only; rotates in 7 days.
Connect Clients: Devices authenticate by channel invite code and PIN.
Chat Securely: All data encrypted client‑side, delivered through volatile memory queues.

Visualize it as a living network that’s constantly regenerating:
relays appear, serve briefly, then disappear without leaving a trace.