Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Jailbreak Detection on OKT
Loading...
1
Correlation Score
Llama-Prompt-Guard-2
0.653368
0.743359
0.83335
0.923341
Feb 14, 2026
Correlation Score
Updated 4d ago
Evaluation Results
Method
Method
Links
Correlation Score
Llama-Prompt-Guard-2
Parameters=86M
2026.02
1
AISA
Backbone=GPT-OSS-20b-I
2026.02
0.9933
AISA
Backbone=Llama3.1-8b-I
2026.02
0.99
AISA
Backbone=Llama2-13b-I
2026.02
0.99
AISA
Backbone=Mistral-7b-I
2026.02
0.98
AISA
Backbone=Qwen3-8b-I
2026.02
0.98
GradSafe
2026.02
0.9733
gpt-4.1-mini
Version=2025-04-14
2026.02
0.97
gpt-5-mini
Version=2025-08-07
2026.02
0.9667
gpt-4o-mini
Version=2024-07-18
2026.02
0.9467
Jailbreak-Classifier
2026.02
0.88
NemoGuard-JailbreakDetect
2026.02
0.8633
SPDetector
2026.02
0.6667
Feedback
Search any
task
Search any
task