Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Safety Detection on MAS LLaMA-3.1-8B (Lookahead)
Loading...
92.31
Accuracy
AgentLens
4.3052
27.1526
50
72.8474
Jun 21, 2026
Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy
AgentLens
Method Category=Interp...
2026.06
92.31
Claude Sonnet-4.6
Method Category=LLM-as...
2026.06
84.62
GPT-4o
Method Category=LLM-as...
2026.06
69.23
Claude Opus-4.6
Method Category=LLM-as...
2026.06
69.23
GPT-5.2
Method Category=LLM-as...
2026.06
61.54
Deepseek-V4-Pro
Method Category=LLM-as...
2026.06
15.38
Gemini-3-Pro
Method Category=LLM-as...
2026.06
7.69
Deepseek-V4-Flash
Method Category=LLM-as...
2026.06
7.69
Feedback
Search any
task
Search any
task