Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Prompt Harmfulness Classification on Public Prompt Harmfulness Benchmarks Suite

81OAI Score

NemoGuard

66.9670.60574.2577.895Jun 26, 2024Oct 17, 2024Feb 8, 2025Jun 1, 2025Sep 23, 2025Jan 14, 2026May 8, 2026
Updated 23d ago

Evaluation Results

MethodLinks
2026.05
81-86.898.575.284.681.6
2026.03
80.657.646.367.15.348.131.7
2026.05
80.5-71.684.457.369.654.3
2026.03
79.26772.280.859.868.451.4
2026.03
79.153.776.488.498.978.977
7925.431.9639.641.8-
2026.03
7925.436.257.69.636.712.1
2026.03
77.374.783.792.999.285.786.1
76.147.171.895.89477-
2026.03
76.147.17689.19475.570.9
75.861.674.19367.274.4-
2026.03
75.861.675.382.567.269.756
74.77382.99970.580-
2026.03
74.772.581.180.370.575.171.5
2026.05
74.1-86.310098.789.488.1
2026.05
73.5-70.69897.282.573
2024.06
72.170.889.499.598.986.1-
2026.05
72.1-80.799.598.98888.9
2026.03
7270.780.894.598.784.288.7
2024.06
70.568.384.410010084.6-
2026.03
70.528.533.647.39.632.34.1
2026.05
69-85.297.79987.787.5
2026.05
68.8-86.199.510088.788.9
2026.03
68.431.93228.54.927.91.8
67.57084.810077.780-
2026.03
67.570.281.277.777.775.578.5