Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Prompt Injection Attack Mitigation on Skill-Inject 139 sandboxes (full set)
Loading...
2.9
ASR
Deepseek-V4-Flash+OC
0.916
14.308
27.7
41.092
Jun 1, 2026
ASR
Updated 1d ago
Evaluation Results
Method
Method
Links
ASR
Deepseek-V4-Flash+OC
Condition=Dynamic
2026.06
2.9
Nemotron3-Super+OC
Condition=Dynamic
2026.06
2.9
Nemotron3-Super+OC
Condition=Static
2026.06
2.9
Deepseek-V4-Flash+OC
Condition=Static
2026.06
5
Sonnet-4.5+CC
Condition=Static
2026.06
7.2
Nemotron3-Super+OC
Condition=Vanilla
2026.06
12.2
Sonnet-4.5+CC
Condition=Dynamic
2026.06
12.9
Sonnet-4.5+CC
Condition=SysTargeted
2026.06
23
Sonnet-4.5+CC
Condition=SysGeneric
2026.06
26.6
Sonnet-4.5+CC
Condition=Vanilla
2026.06
36
Deepseek-V4-Flash+OC
Condition=Vanilla
2026.06
52.5
Feedback
Search any
task
Search any
task