Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Prompt Injection Discovery on AgentDojo
Loading...
84
Source Diversity
PI-Hunter
8.08
27.79
47.5
67.21
Jun 10, 2026
Source Diversity
Instruction Diversity
Updated 1mo ago
Evaluation Results
Method
Method
Links
Source Diversity
Instruction Diversity
PI-Hunter
Model=Gemini-3.1-pro
2026.06
84
82
PI-Hunter
Model=Claude-4.6-sonnet
2026.06
76
62
PI-Hunter
Model=GPT-5.4-mini
2026.06
69
73
Baseline
Model=Claude-4.6-sonnet
2026.06
22
31
Baseline
Model=GPT-5.4-mini
2026.06
18
21
Baseline
Model=Gemini-3.1-pro
2026.06
11
30
Feedback
Search any
task
Search any
task
