Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Harmfulness Evaluation on AgentHarm
Loading...
62.42
HS Score
ReAct
61.68
66.675
71.67
76.665
Jun 4, 2026
HS Score
Harm Score
Updated 1mo ago
Evaluation Results
Method
Method
Links
HS Score
Harm Score
ReAct
Agent Backbone=Average...
2026.06
62.42
47.83
TRIAD + Qwen3.5-9B
Agent Backbone=Average...
2026.06
73.53
9.9
TRIAD + Tri-Guard
Agent Backbone=Average...
2026.06
80.92
13.05
Feedback
Search any
task
Search any
task