Operations console
Your model traffic, at a glance.
Choose the live model, copy a ready-to-use endpoint, and inspect every request without leaving the control room.
Auto-refreshing
Default Model
Loading...
Fallback for GPT requests
Live Selected Model
Loading...
Waiting for client selection
Reasoning Effort
Loading...
Default thinking level
Requests & Tools
0
reqs
0 stream | 0 tools
Tokens Processed
0
In: 0 | Out: 0
Performance
0 ms
0 t/s
Avg latency & tokens/sec
Cursor Integration Endpoints
Public Base URL (Use in Cursor Settings)
checking…
Paste in Cursor Settings → Models → Override OpenAI Base URL
Local Base URL
Direct HTTP endpoint for local scripts or curl requests
Settings & Model Defaults
Loading the selected model's native controls…
Not configured
Not configured
Controls come from the selected model's live capability metadata. Unsupported controls are omitted.
Live Model Playground
Ready
Output will appear here...
Real-time Request Inspector
Auto-updates every 3s| Time | Requested Model | Resolved Model | Type | Tokens (In / Out) | Speed | Latency |
|---|---|---|---|---|---|---|
| No requests recorded yet | ||||||