[ Hub ]
Hub operates the fleet from one place.
One control plane for every node. Deploy, monitor, and manage your entire fleet from a single interface. No SSH. No spreadsheets.
Nodes
4
Serving
3
Fleet Uptime
99.9%
OOM Events
0
Active Deployments
ground-station-08
Svalbard, NorwayVRAM
16.8/24
tok/s
7.8
Uptime
22d
ops-center-03
Edwards AFB, CAVRAM
5.2/16
tok/s
24.1
Uptime
8d
rig-platform-11
North Sea, OffshoreVRAM
6.1/8
tok/s
18.6
Uptime
45d
datacenter-sg-02
Singapore, APAC WarmingVRAM
--
tok/s
--
Uptime
0s
Activity
[ Capabilities ]
What Hub does.
Fig 3.1
Live fleet view
Every node, every model, every region. GPU, memory, throughput, latency, and power, refreshed every second. Thirty days of history by default.
Fig 3.2
Deploy, hot-swap, rollback
Push a model to one machine or the whole fleet. Canary nodes, health checks, automatic rollback on failure. No SSH into individual machines.
Fig 3.3
Identity and audit
SAML and OIDC out of the box. Hub roles map to your IdP groups. Every action logged. Prompt and response content is never captured.
Fig 3.4
Benchmarks and reporting
Throughput, latency, and cost measured across the fleet. Exportable scorecards for your security team or procurement review.
[ Architecture ]
Outbound only. Always.
Every Runtime opens the connection. Hub never reaches into your network. Works behind NAT, on dynamic IPs, and in environments where inbound is impossible.
Fig 4.1 · Outbound topology
Outbound only
Runtime opens the connection. Hub never reaches in. Works through NAT, dynamic IPs, and one-way networks.
Survives disconnect
If the link drops, Runtime keeps serving with its last-known config. State syncs back when Hub returns.
Manage your fleet from one place.
Self-hosted or managed. SSO, audit, rollbacks, and live telemetry on every node.