PIPEROLL / SEISMOGRAPH
Model

glm-5.2@deepinfra

Measured against its own past on the daily and weekly batteries. Each battery has its own history and assessment.

provider channelopenai_compat
identityalias · a floating name the provider may re-point; a change here means what is served under this name changed
daily battery79 readings (2026-08-23 to 2026-11-09)
Latest assessment: Assessed (13/13 metrics eligible)
weekly battery13 readings (2026-08-27 to 2026-11-08)
Latest assessment: Assessed (13/13 metrics eligible)

Published findings

Instruction following pass rate fell Watch

100%→94%
weekly battery · Nominally significant (p<0.01), unconfirmed · first flagged 2026-11-08

Median latency rose Context

3.27 s→6.65 s
weekly battery · Relative p50 shift past threshold, operational context (never a standalone drift claim) · first flagged 2026-11-08

Explore

Open the daily dossier · Open the weekly dossier. Each dossier shows this model and battery, with the full dimension history.