Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Prompt Injection Detection on SafeGuardPI
Loading...
93
F1 Score
IBM Granite Guardian
13.96
34.48
55
75.52
Dec 23, 2025
F1 Score
Updated 4d ago
Evaluation Results
Method
Method
Links
F1 Score
IBM Granite Guardian
Version=3.2, Model Siz...
2025.12
93
IBM Granite Guardian
Version=3.1, Model Siz...
2025.12
92
IBM Granite Guardian
Version=3.2, Model Siz...
2025.12
92
IBM Granite Guardian
Version=3.3, Model Siz...
2025.12
90
IBM Granite Guardian
Version=3.3, Model Siz...
2025.12
86
Llama Guard
Version=3, Reasoning=f...
2025.12
77
Llama Guard
Version=2, Reasoning=f...
2025.12
74
Apriel Guard
Model Size=8B, Reasoni...
2025.12
73
Apriel Guard
Model Size=8B, Reasoni...
2025.12
73
Llama Guard
Version=4, Reasoning=f...
2025.12
70
Llama Prompt Guard 2
Model Size=0.086B, Rea...
2025.12
68
gpt-oss-safeguard
Model Size=20B, Reason...
2025.12
53
Qwen3Guard
Model Size=8B, Constra...
2025.12
37
Qwen3Guard
Model Size=8B, Constra...
2025.12
28
ShieldGemma
Model Size=9B, Reasoni...
2025.12
17
Feedback
Search any
task
Search any
task