Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Red-teaming on HarmBench Claude-Sonnet-3.5 (held-out test)
Loading...
60
ASR
AGENTICRED
10.08
23.04
36
48.96
Jan 20, 2026
ASR
Updated 4d ago
Evaluation Results
Method
Method
Links
ASR
AGENTICRED
2026.01
60
CoP
2026.01
38
AdvReasoning
2026.01
36
J2
2026.01
35
AutoDAN-Turbo
2026.01
12
Feedback
Search any
task
Search any
task