Agents (Healthy / Total)
–
Tunnels (Connected / Agents)
–
Models (unique)
–
Queue Depth (sum)
–
CPU (avg %)
–
Memory (avg MB)
–
GPUs (total)
–
VRAM (used/total)
–
Throughput (sum tps)
–
vLLM Instances
–
vLLM Tokens
–
vLLM Cache Hit
–
vLLM KV Cache
–
Inference Pools (/inference/{pool} — tunnel-served) [+ new pool]
| Pool | Models | Members | Config | Requests | Failovers |
|---|
Live Request Feed [pause] [clear]
Models
| Model | Agents | Capabilities |
|---|
vLLM Instance Metrics [hide]
| Instance | Agent | Run | Wait | Prompt Tokens | Gen Tokens | Cache Hit | KV Cache | GPU Cache | Avg TTFT | Avg ITL | Req (S/L/A) | TTFT p50 | TTFT p95 | TTFT p99 |
|---|