Hero summary
Cluster health
Total servers
6
5 online
Active GPUs
7
8 GPUs registered
Running models
5
Deployed inference endpoints
Queued jobs
4
Waiting in scheduler
Tokens per second
892
Fleet aggregate throughput
Power usage
2.8 kW
Estimated live draw
Tokens per second
VRAM usage
Power draw
Job completion time
Active alerts
5 alertsActivity log
terminalServer fleet
searchable / sortable| Hostname | Status | CPU | RAM | Disk | Network | Uptime | Assigned models |
|---|
GPU monitor
live telemetryModel deployments
sortable| Model | Quantization | Context | GPU assignment | Status | Jobs | Tokens/s |
|---|
Inference queue
pause / resume| Job ID | Model | Server | Status | Progress | Tokens | ETA | Priority | Action |
|---|