benchmaxxer Sep 23, 2026
U18Browse 296

U18

Does the model follow the publisher's safety policy under challenging prompts?

Version not specifiedExact reported variant
Safety3 ranked models5 reported values1 reportsHigher is better

Top rankings

One row per model, using its best reported score across effort settings.

3
RankModelBest scoreBest reported setting
1GPT-5.6 SolOpenAI0.865Reported configurationOpenAI report1 value · 1 reportAug 6, 2026 · source
2GPT-5.6 LunaOpenAI0.857Reported configurationOpenAI report1 value · 1 reportAug 6, 2026 · source
3GPT-5.5OpenAI0.756Reported configurationOpenAI report3 values · 1 reportAug 6, 2026 · source

Effort curve

Every sourced cost-linked effort value for this exact version. Lines connect complete sweeps only.

0
No cost-linked effort sweep for this version.
Definition and comparison boundarylimited methodology

Does the model follow the publisher's safety policy under challenging prompts?

Publisher-defined safety behavior, robustness, or preparedness evaluations under the stated test protocol. Higher is better. Raw not_unsafe share in the publisher's [0,1] scale.

Harness, tools, prompts, attempt count, and benchmark version can materially change the result. Compare within one reporting context.

Dedicated U18 evaluation on difficult production-derived examples; scores are not estimates of typical production prevalence.

Method / source