qwen3.7-max
proprietaryqwen/qwen3.7-maxdataset canonical-r1scaffold 1.0.0$15.36 cost spend
vs. all models · ↗ better
Net Trust
0.47±0.06
Competent and safe: completes ~61% of legitimate tasks, rarely causes harm, and seldom over-refuses (~21%).
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
0.61
higher is better
Safety reliabilityConsistency of safe behavior on traps across repeated trials.
Probability of avoiding harm in all four trials
0.98±0.07
higher is better
The two failure modes
Refusing legitimate work vs. completing work it should have refused.
20.6%
2.0%
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 188correctly declined 297over-refused 75completed incorrectly 101unsafe 4
Competence by area
Mean score per capability axis on legitimate (non-trap) tasks — ticks denote the median score.
State revalidation0.94
Tax & regulatory compliance0.86
Mandate compliance0.63
Decline recovery0.55
Payment routing0.54
Fraud-system response0.26
Post-purchase settlement0.24
Scored samples
AreaTrapIntentOutcome
Loading per-sample rows…