Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Denial of Service (DoS) attack on ToolSafe (test)
Loading...
61,981
Think Length
beam-search optimization framework
-415.88
15,783.31
31,982.5
48,181.69
Jun 12, 2026
Think Length
Ratio
Updated 1mo ago
Evaluation Results
Method
Method
Links
Think Length
Ratio
beam-search optimization framework
Target Model=Claude-3....
2026.06
61,981
27.7
ReasoningBomb
Target Model=Claude-3....
2026.06
2,379
1.2
RECUR†
Target Model=Claude-3....
2026.06
2,298
1.16
OverThink
Target Model=Claude-3....
2026.06
2,283
1.15
CRABS
Target Model=Claude-3....
2026.06
2,270
1.14
ENGORGIO
Target Model=Claude-3....
2026.06
2,207
1.11
ThinkTrap†
Target Model=Claude-3....
2026.06
2,207
1.11
Length-matched benign
Target Model=Claude-3....
2026.06
1,984
1
Feedback
Search any
task
Search any
task