Helicone history

Aug 2026 → Nov 2026

openai → openai cell. The engine changed between these runs (f3adbb1 → d240bad), so differences can include instrument changes, not only gateway changes. Changes within ±5% show as roughly unchanged (—); this is a display tolerance, not a statistical margin.

ReadingAug 2026Nov 2026ChangeRank Aug 2026 → Nov 2026
Added latency p50 (µs)274 µs256 µs↑ -6.6%7 of 14 → 7 of 15
Added latency p99 (µs)294 µs274 µs↑ -6.8%5 of 14 → 4 of 15
Throughput, p99 ≤ 1 ms (req/s)14,073 req/s15,405 req/s↑ +9.5%3 of 3 → 2 of 2
Throughput, p99 ≤ 5 ms (req/s)14,865 req/s16,288 req/s↑ +9.6%2 of 2 → 2 of 2
Throughput, p99 ≤ 10 ms (req/s)14,941 req/s16,376 req/s↑ +9.6%5 of 5 → 3 of 3
Throughput, p99 unbounded (req/s)14,941 req/s16,376 req/s↑ +9.6%4 of 5 → 1 of 2
CPU per request (µs)276.8259.4↑ -6.3%7 of 14 → 6 of 14
Idle RSS (MiB)42.9 MiB42.97 MiB— +0.2%— → —
Peak RSS (MiB)56.6 MiB56.95 MiB— +0.6%— → —

Runs

2 published runs · Added latency p99 (µs), openai → openai · values describe the recorded local-mock configuration; a missing reading is never zero. Hover a trend for its comparability details.

Added latency p99 (µs), openai → openai
MeasuredBuildEngineResultTrendField placementEvidence
target/release/ai-gatewayd240bad274 µsEngine changed4 of 15 Current fieldRun evidence
target/release/ai-gatewayf3adbb1294 µsFirst run5 of 14 Aug 2026 fieldRun evidence
Compare two runs across every recorded metric

History JSON · All gateways