Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Safety Evaluation on Malicious Instruct v1 (test)
Loading...
45
Compliance Rate
DeepSeek-R1-Distill-Qwen-7B
-1.8
10.35
22.5
34.65
Jul 14, 2025
Compliance Rate
Updated 18d ago
Evaluation Results
Method
Method
Links
Compliance Rate
DeepSeek-R1-Distill-Qwen-7B
Model Category=Local,...
2025.07
45
Qwen2.5-7B-Instruct
Model Category=Local,...
2025.07
7
o1
Model Category=API-bas...
2025.07
1
o3-mini
Model Category=API-bas...
2025.07
1
STAIR-SFT
Model Category=Local,...
2025.07
0
STAIR-DPO
Model Category=Local,...
2025.07
0
STAR-1
Model Category=Local,...
2025.07
0
ARMOR
Model Category=Local,...
2025.07
0
ARMOR-Think
Model Category=Local,...
2025.07
0
Feedback
Search any
task
Search any
task