Network latency rollup derived purely from indexed receipts' signed `startedAt`/`completedAt` pairs — the last ≤100 per group (overall, per-provider, per-model). No side metrics store: the receipts are the source of truth, so the numbers cannot drift from them.
Output
Encoding
application/jsonbyModel
unknown
Optional
No description available.
byProvider
unknown
Optional
No description available.
overall
unknown
Optional
No description available.
Try It
Requests are sent directly from your browser. Some servers may block requests due to CORS.
View raw schema
{
"type": "query",
"output": {
"schema": {
"type": "object",
"properties": {
"byModel": {
"type": "unknown"
},
"overall": {
"type": "unknown"
},
"byProvider": {
"type": "unknown"
}
},
"description": "Latency overview, grouped overall and by provider and model. Each group reports a sample size and summary statistics (e.g. p50/p95) over the receipts' start→complete durations."
},
"encoding": "application/json"
},
"description": "Network latency rollup derived purely from indexed receipts' signed `startedAt`/`completedAt` pairs — the last ≤100 per group (overall, per-provider, per-model). No side metrics store: the receipts are the source of truth, so the numbers cannot drift from them."
}