Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Harmful content detection on CONAN
Loading...
98.47
Accuracy
LlamaGuard-1
86.1772
89.3686
92.56
95.7514
Apr 18, 2026
Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy
LlamaGuard-1
2026.04
98.47
LlamaGuard-1
2026.04
98.47
Perspective API
2026.04
96.97
Perspective API
2026.04
96.97
OpenAI Moderation
2026.04
95.29
OpenAI Moderation
2026.04
95.29
LlamaGuard-2
2026.04
86.65
LlamaGuard-2
2026.04
86.65
Feedback
Search any
task
Search any
task