Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
LLM Safety and Informativeness Evaluation on BeaverTails
Loading...
87.2
Safety Rate
SafeMoE-XL
12.84
32.145
51.45
70.755
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
87.2
6.4
Oyster-I 14B
Parameters=14B
2026.05
81.7
7.6
RealSafe-R1
2026.05
68.8
5.6
SafeMoE-Qwen
Backbone=Qwen
2026.05
63.4
6.9
Deepseek-R1-Qwen
Backbone=Qwen
2026.05
53.6
6.8
SN-Tune
2026.05
51.6
5.2
Zephyr
2026.05
39.8
5.9
Qwen-3B
Parameters=3B
2026.05
34
6.2
Mistral-7B
Parameters=7B
2026.05
31.4
3.9
Mistral-SFT
2026.05
22.4
5.9
SafeLoRA
2026.05
15.7
4.5
Feedback
Search any
task
Search any
task