Digest · week 2026-W46

The week, counted.

This week's reading

7 of 20 endpoints checked this week have a daily serving check that differs from its stored reference. This is a signal to look at, not a finding about any model. 18 endpoints have a battery candidate awaiting the battery of 29 November 2026. The behavioral battery recorded no movement past the public line.

100 canary checks run (86 matched baseline) · 5000 battery trials across 20 seats · 973 stances held · 0 seats not measured. Pre-series · generated 2026-11-18

A change in when this reading is taken

The weekly series changed when it is measured. Until 30 August 2026 the weekly battery was taken on Monday morning. From that date it is taken on the Sunday night that ends the week it is labelled with. The reading is the same measurement of the same items on the same roster; only the instant it is taken has moved, by about 32 hours. This bureau has not measured whether the time a reading is taken affects a stance reading, so week-over-week comparisons that straddle this date are not treated as evidence of change, and no correction has been applied because none is known to be needed.

THE TRANSITION ALSO COST A READING. The old schedule's next run was removed nine hours before it would have fired and the new schedule's first slot had already passed, so no battery was taken in that week and none has been manufactured since. The readings either side of the change are 12.88 days apart rather than about seven. Their week labels are consecutive because a label names when a reading was taken, so the labels alone will not show you this and the interval is published beside them for that reason.

Canary checks
86of 100 match
Battery trials
5000
Stances held
973
Material behavioral changes
0battery candidate

Actions this occasion

one state per seat, silence quantified
2026-11-18 7 of 20 endpoints checked this week have a daily serving check that differs from its stored reference. This is a signal to look at, not a finding about any model. 18 endpoints have a battery candidate awaiting the battery of 29 November 2026.

Two different things are watched here. Serving checks (daily) ask whether the endpoint you call is still the build we baselined. Behavioral measurements (every second Sunday) ask whether its positions moved. A movement measured past the line on two batteries in a row, a check that is not yet built, is the only thing we ever call a material behavioral change. None has been published.

Serving check differs this week anthropic-a ›
Serving check against its stored reference
Battery candidate, awaiting the next battery · 3 questions
1 serving check differed today
Serving check matches its reference openai-a ›
Serving check against its stored reference
Battery candidate, awaiting the next battery · 2 questions
detects ≥ 1.94
Serving check matches its reference xai-a ›
Serving check against its stored reference
Battery candidate, awaiting the next battery · 6 questions
detects ≥ 3.17
Serving check matches its reference deepseek-a ›
Serving check against its stored reference
Battery candidate, awaiting the next battery · 4 questions
detects ≥ 2.51
Serving check matches its reference anthropic-b ›
Serving check against its stored reference
Battery candidate, awaiting the next battery · 1 question
detects ≥ 1.94
Serving check matches its reference anthropic-c ›
Serving check against its stored reference
Battery candidate, awaiting the next battery · 1 question
detects ≥ 1.58
Serving check differs this week anthropic-d ›
Serving check against its stored reference
Battery candidate, awaiting the next battery · 4 questions
2 serving checks differed today
Serving check matches its reference anthropic-e ›
Serving check against its stored reference
Battery candidate, awaiting the next battery · 2 questions
detects ≥ 1.94
Serving check matches its reference openai-b ›
Serving check against its stored reference
Battery candidate, awaiting the next battery · 1 question
detects ≥ 2.51
Serving check matches its reference openai-c ›
Serving check against its stored reference
Battery candidate, awaiting the next battery · 2 questions
detects ≥ 2.69
Serving check differs, below the line google-b ›
Serving check against its stored reference
Battery candidate, awaiting the next battery · 3 questions
1 serving check differed today
Serving check matches its reference deepseek-b ›
Serving check against its stored reference
Battery candidate, awaiting the next battery · 3 questions
detects ≥ 1.94
Serving check matches its reference openweight-a ›
Serving check against its stored reference
detects ≥ 2.38
Serving check differs this week openweight-b ›
Serving check against its stored reference
Battery candidate, awaiting the next battery · 6 questions
1 serving check differed today
Serving check differs this week openweight-c ›
Serving check against its stored reference
Battery candidate, awaiting the next battery · 5 questions
2 serving checks differed today
Serving check differs, below the line openweight-d ›
Serving check against its stored reference
Battery candidate, awaiting the next battery · 11 questions
2 serving checks differed today
Serving check matches its reference openweight-e ›
Serving check against its stored reference
Battery candidate, awaiting the next battery · 7 questions
detects ≥ 2.38
Serving check matches its reference openweight-f ›
Serving check against its stored reference
Battery candidate, awaiting the next battery · 5 questions
detects ≥ 2.38
Serving check differs this week openweight-g ›
Serving check against its stored reference
Battery candidate, awaiting the next battery · 6 questions
1 serving check differed today
Serving check matches its reference simulant-a ›
Serving check against its stored reference
detects ≥ 3.63

Not measured: google-a, google-c, google-d, xai-b.

One serving state per endpoint each day, in the daily serving-check vocabulary. An endpoint whose serving check matches its reference ships the smallest shift this reading could detect (n and power pre-registered); a new serving baseline means the served build changed and comparisons reset. A battery candidate is shown on its own line and is not a serving state. A material behavioral change is only ever declared by the behavioral battery, never by a daily serving check. actions.json · how states are decided

Reading the quiet

Nothing moved past the pre-registered thresholds this week, and that sentence is the product. A change-alert service earns the right to be believed on its loud days by publishing its quiet ones: same cadence, same thresholds, same evidence trail, week after week. When a change does confirm, it appears here, on the changelog, and as a diff card with the full evidence bundle.

Pre-series note: this digest counts pipeline activity while the instrument is still measuring itself; no findings publish until methodology v1.0 freezes.

Want this scoped to your builds, “all quiet on the 3 builds you watch”? That is Watch, the recurring product, in preparation.

About Watch →
Cite this Permalink https://modelometer.com/digest · Run hash ffb15bb915bba5ff81b0764a65643f24aeecf3c3428d63189bff6b125a6a9d8e · Retrieved 2026-11-18 · week 2026-W46: the behavioral battery recorded no movement past the public line across 5000 trials (pre-series).