SpliceVPN builds an encrypted mesh across all your endpoints — each one a single machine or a whole network — then bonds every internet link at each node for more bandwidth, more resilience, or both. It finds a path between any two nodes on its own, and when there's no other way, it falls back to a relay we run for you.
Everything below happens automatically once the mesh is up. You don't configure routes, exchange keys by hand, or touch a firewall rule.
The current MVP creates a controlled hub-relayed WireGuard overlay between endpoints. A lightweight coordinator helps nodes find each other; traffic is relayed through your own SpliceVPN hub in the current MVP.
Each node can be a single computer or an entire site. Bridge offices, attach a laptop to a LAN, or join networks that were never meant to meet.
Where you have more than one internet connection, SpliceVPN combines them — for aggregate bandwidth, automatic failover, or both at once.
The current MVP uses your own SpliceVPN hub relay so nodes can communicate without opening inbound client firewall rules. Direct traversal and hole-punching are planned for a later build.
UDP, TCP, or both, across one port or many. SpliceVPN adapts to whatever the network in the middle is willing to let through.
Run the software, answer a few questions about what you want, and the rest is handled — discovery, keys, routing, failover, all of it.
SpliceVPN works out connectivity for you, between every pair of nodes, wherever they happen to be. It climbs down a ladder until something connects — and the bottom rung is a relay you don't have to find or maintain, because we provide it.
If a reachable path exists between two nodes, SpliceVPN uses it. Lowest latency, no middlemen, full bandwidth.
Behind NAT or a strict firewall, SpliceVPN negotiates a path through — hole-punching and multiplexing over whatever ports are open.
Current MVP traffic routes through your own SpliceVPN hub relay. Direct and traversal modes are planned for a later build.
No manual key exchange, no routing tables, no firewall archaeology. SpliceVPN asks what matters and works out the rest.
A single machine or a whole network — add as many nodes as you like.
More speed, more reliability, or a balance of the two across your links.
Specific ports or protocols only? Otherwise SpliceVPN uses whatever works.
No manual keys, no config files. On Windows it's a single click — the installer sets up WireGuard and the agent for you.
Windows — download the one-click installer and run it. It installs WireGuard and the SpliceVPN agent for you; just approve the admin prompt.
Linux — download the build, then tar xzf splicevpn.tar.gz && sudo ./splicevpn/bin/splice-agent
Paste the join key and coordinator address your administrator gives you. That's the only input the agent needs — no key exchange, no routing tables.
The agent discovers every other node, brings up the encrypted tunnel, and keeps it connected — automatically, even after a reboot. Nothing else to configure.
Install SpliceVPN on each node, answer three questions, and let it find the way. We'll be the safety net.
or one-line install on Windows — PowerShell as Administrator (installs WireGuard for you):
iwr https://splicevpn.com.au/install-splicevpn.ps1 -OutFile $env:TEMP\splice.ps1; Start-Process powershell "-ExecutionPolicy Bypass -File $env:TEMP\splice.ps1" -Verb RunAs