Home › Evidence › Records › eng-0118
⚠ This notebook has stopped taking notes: newest record is 23 days old (eng-0280), against an expected cadence of 14 days.

eng-0118

window recorded
citable URL: https://halobench.com/records/eng-0118/ — this address never moves; the anchor /records/#eng-0118 keeps resolving
25.70 Whmean 114.5 W − 10.1 W idle floor (box-idle-all-empty) = 104.4 W attributable
window 2026-08-16T07:25:32Z → 2026-08-16T07:39:00Z · provenance recorded
method wall-meter · aihydra energy sensor via Home Assistant (counter-difference on sensor.hardware_ai_hydra_energy, raw ~10s-resolution history samples, linearly interpolated to the run-meta window) · no contention
covers run-0319 run-0320
aihydra laguna-s21-fullbench throughput matrix cell rocm-d32768 (3 fresh-process reps, pp1024/tg256 at depth 32768). Counter interpolated from raw ~10s-resolution HA history samples pulled fresh this session: start 21.650084960847323 kWh, end 21.675783821154234 kWh. Recomputed independently from HA raw history; matches the MODEL-PAGE.md draft to within rounding (25.70 Wh / 114.5 W).

Wh, never joules — read as a cumulative-counter difference across the window; delta over the idle baseline is the only figure that means anything

Cited by — computed at build time, never stored

model pages laguna-s-21