end-to-end target for latency-sensitive inference; ~100 ms is the human 'instantaneous' threshold
sub-50 msforecast
| Value kind | forecast — Forecasts are time-bound outlooks, not specifications; test decisions across plausible scenarios. |
|---|---|
| As of | 2026 |
| Source | HumAI; practitioner edge-inference benchmarks · Edge-inference latency discussion; under 50 ms is identified as the target for latency-sensitive inference, with approximately 100 ms described as the human instantaneous-response threshold. |
| Review | checking…review by 2026-09-06 · standard cadence |
| Recorded changes | last 2026-06-29 |
| Claim id | end-to-end-target-for-latency-sensitive |
Where the guide uses it
Not quoted in a chapter yet; it is kept in the curated register.
← Full numbers register — every date-stamped figure in the guide, with revision history.