Models ready
workers reporting ready
Requests
all time (local usage store)
Avg latency
across all ops
Errors
HTTP ≥ 400
Models
Status and headline benchmark metric per model. Each model runs in its own worker process.
| Model | Task | Version | Status | Benchmark |
|---|---|---|---|---|
Requests · last 48 h
Per hour, all tasks
No requests yet — try a playground.
Recent requests
Audit rows recorded by the gateway (no payloads, only metering).
| Time | Task · op | Model | Usage | Latency | Status |
|---|---|---|---|---|---|
| Nothing yet. | |||||