List your company

Safety Judge Evaluation on High-Confidence (HC) (test)

99.9F1 ScoreClaude
11.81234.68157.5580.419Jun 24, 2026

Evaluation results

MethodLinks
Claude2026.0699.9—10099.799.799.9
SorryBench2026.0698.2—10096.596.698.2
AILuminate2026.0696.8—10093.79496.9
StrongReject2026.0696.4—10093.196.399.7
Ettin-400MSubset=HQ2026.069088.790.489.692.298.4
Ettin-150MSubset=HQ2026.0685.383.4908188.397.8
Ettin-400MSubset=NF2026.0685.183.298.4759699.3
JBB2026.0683.5—10071.672.885.8
Ettin-150MSubset=NF2026.068078.195.668.89398.8
LG32026.0675.2—10060.361.980.1
LG42026.0649.9—10033.23666.6
ShieldGemma-9B2026.0616.315.19.848.68.973.7
ShieldGemma-2B2026.0615.213.510.825.68.467.5