Two changes on 11 September: the acknowledgement stopped blocking the provider call, then only task-relevant tools were offered to the model. Median turn time fell from 3.08 s to 1.27 s.
Two changes on 11 September: the acknowledgement stopped blocking the provider call, then only task-relevant tools were offered to the model. Median turn time fell from 3.08 s to 1.27 s.
Test: Latency Lab (tools/latency/lab.py): 21 scripted journeys (status, time, recall, project, hud, latency, chat; three passes) through the normal launcher; end of input to end of turn.
Protocol
Latency Lab (tools/latency/lab.py): 21 scripted journeys (status, time, recall, project, hud, latency, chat; three passes) through the normal launcher; end of input to end of turn.
Limits
Log
| Variant | Value | Evidence | Samples | Source |
|---|---|---|---|---|
| Baseline, pre-optimisation 11 Sept 2026 | 3,083 ms | MEASURED | 21 | docs/analysis/latency/lab-20260911T221912Z.json |
| Acknowledgement no longer blocks provider call 11 Sept 2026 | 1,723 ms | MEASURED | 21 | docs/analysis/latency/lab-20260911T222249Z.json |
| Plus task-relevant tool projection 11 Sept 2026 | 1,270 ms | MEASURED | 21 | docs/analysis/latency/lab-20260911T223508Z.json |
Source paths refer to the private JARVIS repository and runtime records. They are listed for audit and are not published.