Real-time fleet overview
Track machine totals, fleet averages, disk capacity, network information, and GPU utilization across online, rented, and offline hosts in one place.
Vast.ai fleet monitoring
Stay on top of GPU hosts from the phone in your pocket.
Vast Monitor is a mobile-friendly monitoring dashboard for Vast.ai GPU hosts with real-time metrics, fleet health, alerting, and earnings tracking across web and Flutter clients.
- Real-time fleet visibility
- Mobile + web clients
- Alerts and earnings history
Vast Monitor dashboard preview
Illustrative product preview showing fleet overview, machine states, GPU load, temperature, power, earnings, and active alerts.
Core capabilities
Built for operators who need signal, not noise.
Vast Monitor brings together machine health, fleet status, and commercial context so you can react quickly whether you are at the rack or away from the keyboard.
Per-machine telemetry
Inspect CPU, RAM, GPU utilization, GPU memory usage, temperature, power draw, disk, and system temperature without digging through separate dashboards.
Alert rules and events
Use alerting workflows to catch thermal spikes, host downtime, or capacity issues before they become missed revenue or support incidents.
Earnings visibility
Keep operational status tied to earnings history so you can understand how fleet health and rental state affect revenue over time.
How it works
Connect your account, stream metrics, monitor the fleet.
- 01
Connect Vast.ai
Link your Vast.ai account to discover hosts, instances, and the current state of your fleet.
- 02
Collect time-series metrics
Ingest telemetry through the Next.js backend and monitoring pipeline powered by VictoriaMetrics and Grafana Alloy.
- 03
Monitor anywhere
Review fleet health, alert events, and earnings from the Flutter mobile app or the web dashboard.
Metric coverage
See the hardware signals that matter for host reliability.
Compute load
- CPU utilization
- RAM usage
- GPU utilization
- GPU memory utilization
Thermals and power
- GPU temperature
- GPU power draw
- System temperature
- Trend-aware averages
Capacity and connectivity
- Disk utilization and capacity
- Network information
- Fleet instance visibility
- Machine state changes
Alerts and earnings
Keep uptime and revenue in the same conversation.
Vast Monitor pairs alert rules and event history with earnings tracking so you can see when a thermal issue, downtime event, or capacity bottleneck starts to affect the business side of the fleet.
Alert rules
Define the thresholds and events that deserve immediate attention.
Earnings history
Review revenue context alongside machine and rental state trends.
Implementation stack
Mobile-first product, backend-backed observability.
The experience spans a Flutter client, a Next.js API backend, and time-series metrics infrastructure designed for operational visibility rather than static reports.
Get updates on Play Store launchReady to explore Vast Monitor?