Safety Monitoring
Benchmarks
DatasetSOTA methodMetricTrendResultsLast updated
89.9Accuracy
40
May 26, 2026
92.5Precision
15
Mar 27, 2026
67.86ASR @ 1% FPR
5
Mar 25, 2026
34.15ASR @ 1% FPR
5
Mar 25, 2026
67.02ASR @ 1% FPR
5
Mar 25, 2026
32.96ASR @ 1% FPR
5
Mar 25, 2026
19.6ASR @ 1% FPR
5
Mar 25, 2026
80.3Safety AUROC
4
Mar 27, 2026
92.9Safety AUROC
4
Mar 27, 2026