Sample Size
50
Endpoint
Exact Match
Mode
Live
End-to-End Latency is the total round-trip time including network overhead.
Service Processing is the time the server reports for the internal search computation.
Additional Time = End-to-End minus Service Processing — this captures network transit and HTTP parsing overhead.
Ideally, Additional Time should be small and consistent across requests.
Request Timing Samples
Latency Breakdown — 50 Sequential Requests
Timing Active
Latency is color-coded: green (<5ms) is fast,
amber (5–20ms) is moderate,
red (>20ms) may indicate load or network issues.
| Req # | Endpoint | End-to-End Latency | Service Processing | Additional Time |
|---|---|---|---|---|
|
Enter an API key and click Run Latency Test to begin
|
||||
Developer Reference
Search Endpoint
This page benchmarks the GET /exact_match/{query} endpoint.
Copy a snippet to integrate the same call into your application and measure latency from your own environment.
The response includes a time field (server-side processing ms) so you can subtract it from your
end-to-end latency to isolate network overhead.
Interpretation Guide
Service < 2ms: The search engine is performing excellently. Most latency is network.
Service 5–20ms: Moderate load or a large index. Consider checking document count and custom field cardinality.
Additional time > 20ms: Possible network congestion, large response payload, or server resource constraints.