Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
LLM Safety and Informativeness Evaluation on AdvBench
Loading...
97.2
Safety Rate
SafeMoE-XL
8.488
31.519
54.55
77.581
May 30, 2026
Safety Rate
Informativeness Score
Updated 1d ago
Evaluation Results
Method
Method
Links
Safety Rate
Informativeness Score
SafeMoE-XL
Variant=XL
2026.05
97.2
8.2
Oyster-I 14B
Parameters=14B
2026.05
80.5
6.9
SafeMoE-Qwen
Backbone=Qwen
2026.05
73.2
7.8
RealSafe-R1
2026.05
60.8
5.8
SN-Tune
2026.05
55.2
6.9
Deepseek-R1-Qwen
Backbone=Qwen
2026.05
48.9
6.7
Zephyr
2026.05
44.5
7.2
Qwen-3B
Parameters=3B
2026.05
31.1
4.4
Mistral-SFT
2026.05
27.7
6.7
SafeLoRA
2026.05
25.1
5.9
Mistral-7B
Parameters=7B
2026.05
11.9
4.6
Feedback
Search any
task
Search any
task