Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Response Harmfulness Detection on BeaverTails
Loading...
89.9
F1 Score
BeaverDam 7B
62.5584
69.6567
76.755
83.8533
Feb 3, 2026
F1 Score
Updated 4d ago
Evaluation Results
Method
Method
Links
F1 Score
BeaverDam 7B
Model Category=LLM Gua...
2026.02
89.9
GuardReasoner 8B
Model Category=LLM Gua...
2026.02
87.6
MD-Judge 7B
Model Category=LLM Gua...
2026.02
86.7
Qwen3Guard Gen 8B
Model Category=LLM Gua...
2026.02
86.48
GuardReasoner-Omni 4B
Model Category=VLM Gua...
2026.02
86.04
GuardReasoner-Omni 2B
Model Category=VLM Gua...
2026.02
85.89
Qwen3Guard Gen 8B
Model Category=LLM Gua...
2026.02
85.35
GuardReasoner-VL 3B
Model Category=VLM Gua...
2026.02
85.19
GuardReasoner-VL 7B
Model Category=VLM Gua...
2026.02
84.99
WildGuard 7B
Model Category=LLM Gua...
2026.02
84.4
PolyGuard-Qwen 7B
Model Category=LLM Gua...
2026.02
79.39
HarmBench LLaMA 13B
Model Category=LLM Gua...
2026.02
77.1
HarmBench Mistral 7B
Model Category=LLM Gua...
2026.02
75.2
Aegis Guard Defensive 7B
Model Category=LLM Gua...
2026.02
74.7
Aegis Guard Permissive 7B
Model Category=LLM Gua...
2026.02
73.8
LLaMA Guard 4 12B
Model Category=VLM Gua...
2026.02
69.51
LLaMA Guard 3 8B
Model Category=LLM Gua...
2026.02
67.84
ShieldGemma 9B
Model Category=LLM Gua...
2026.02
63.61
Feedback
Search any
task
Search any
task