One self-hosted stack. Full visibility, full control.
Real-time metrics
CPU, RAM, disk, network and load — updated every 4 seconds, 30 days of history stored locally.
4s refresh30-day history
Multi-GPU monitoring
Utilization, VRAM, temperature and power draw across every GPU — simultaneously.
NVIDIAMulti-GPU
Auto-discovery
Automatically detects new servers, containers and services as they come online. No manual setup.
Zero-configAuto-register
AI assistant
Ask anything about your infrastructure in plain language. Runs on-device with full context — no cloud.
On-deviceAI agents
Container monitoring
Resource usage, health status and live logs for every container. See how your stack is connected.
DockerLive logs
Security monitoring
Detect intrusion attempts, audit SSH access and flag suspicious processes — built in, no extra tools.
Intrusion detectionSSH audit
Alerts & notifications
Set thresholds for any metric. Get notified the moment something needs attention.
Multi-channelAlert history
Integrations
REST API and webhooks let you pipe metrics and alerts into any external tool or workflow.
REST APIWebhooks
Privacy by design
No telemetry, no third-party access. Your data stays inside your network, period.
On-premiseNo telemetry