Enter the ADMIN_TOKEN configured in Coolify.
Models
Loaded
Active slots
24h requests
AI model dashboard
Updates every 5 seconds · model counters reset when the model restarts
Portal uptime
Generation throughput · recent samples
Server-reported average tokens/second, not an instantaneous rate.
Waiting for metrics
Inference slots
| Slot | Status | Context capacity |
|---|---|---|
GPU utilization and VRAM are not exposed by this metrics endpoint.
Models
Completion is confirmed by the server after all files finish.
| Model | Status | Architecture | Actions |
|---|---|---|---|
|
|
|||
| No cached models yet. | |||
API Keys
Copy this key now. It will not be shown again.
| Name | Prefix | Requests | |
|---|---|---|---|
Monitoring
Slots
Active
Avg latency 24h
Errors 24h
Raw llama.cpp metrics
Recent Requests
| Time | Key | Model | Path | Status | Duration |
|---|---|---|---|---|---|