Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Attack Success Rate Evaluation on Hot100
Loading...
26
ASR
SkillAttack
0
6.75
13.5
20.25
Apr 5, 2026
ASR
Updated 11d ago
Evaluation Results
Method
Method
Links
ASR
SkillAttack
Model=glm-5
2026.04
26
SkillAttack
Model=gpt-5.4
2026.04
23
SkillAttack
Model=hunyuan-2.0-inst...
2026.04
23
SkillAttack
Model=doubao-seed-1-8-...
2026.04
19
SkillAttack
Model=gemini-3.0-pro-p...
2026.04
16
SkillAttack
Model=kimi-k2.5
2026.04
15
SkillAttack
Model=gpt-5.4-nano
2026.04
12
SkillAttack
Model=Qwen3.5-Plus
2026.04
11
SkillAttack
Model=claude-sonnet-4-...
2026.04
10
SkillAttack
Model=MiniMax-M2.5
2026.04
9
Direct Attack
Model=hunyuan-2.0-inst...
2026.04
4
Direct Attack
Model=glm-5
2026.04
3
Direct Attack
Model=gpt-5.4
2026.04
2
Direct Attack
Model=kimi-k2.5
2026.04
2
Direct Attack
Model=Qwen3.5-Plus
2026.04
2
Direct Attack
Model=MiniMax-M2.5
2026.04
2
Direct Attack
Model=doubao-seed-1-8-...
2026.04
2
Direct Attack
Model=gpt-5.4-nano
2026.04
1
Direct Attack
Model=claude-sonnet-4-...
2026.04
1
Direct Attack
Model=gemini-3.0-pro-p...
2026.04
1
Feedback
Search any
task
Search any
task