glm-5
open-weightsz-ai/glm-5dataset canonical-r1scaffold 1.0.0$8.75 cost spend
vs. all models · ↗ better
Net Trust
0.55±0.06
Executes payments it should refuse: unsafe completions in ~12% of trap scenarios. Competent (69%) but not reliably safe.
At a glance
Competence and safety reliability, each with a tick at the all-models median.
CompetencePerformance on legitimate (non-trap) work.
Mean score on legitimate tasks
0.69
higher is better
Safety reliabilityConsistency of safe behavior on traps across repeated trials.
Probability of avoiding harm in all four trials
0.92±0.09
higher is better
The two failure modes
Refusing legitimate work vs. completing work it should have refused.
9.9%
11.6%
over-refuses (False-refusal rate)Over-caution — legitimate work the model declined.over-completes (Unsafe-completion rate)Payments executed when the correct action was to abort or escalate.
No catastrophic failures observed.
Outcomes across 665 scenarios
Outcome classifications only; scenario content and gold answers remain private to protect benchmark integrity.
completed correctly 228correctly declined 279over-refused 36completed incorrectly 103unsafe 19
Competence by area
Mean score per capability axis on legitimate (non-trap) tasks — ticks denote the median score.
State revalidation0.95
Tax & regulatory compliance0.93
Mandate compliance0.70
Payment routing0.53
Post-purchase settlement0.35
Fraud-system response0.34
Decline recovery0.11
Scored samples
AreaTrapIntentOutcome
Loading per-sample rows…