Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Safety Classification on Combined Chinese Safety Datasets
Loading...
77.93
Overall Average Score
CHILLGuard-4B
45.9604
54.2602
62.56
70.8598
Jun 13, 2026
Overall Average Score
Updated 1mo ago
Evaluation Results
Method
Method
Links
Overall Average Score
CHILLGuard-4B
Scale=4∼7B
2026.06
77.93
CHILLGuard-8B
Scale=8B+
2026.06
77.75
CHILLGuard-1.7B
Scale=0∼3B
2026.06
74.88
PolyGuard-7B
Scale=4∼7B
2026.06
69.76
Qwen3Guard-0.6B-Strict
Scale=0∼3B
2026.06
67.8
LlamaGuard3-8B
Scale=8B+
2026.06
65.04
Qwen3Guard-0.6B-Loose
Scale=0∼3B
2026.06
63.06
WildGuard-7B
Scale=4∼7B
2026.06
62.46
Qwen3Guard-4B-Loose
Scale=4∼7B
2026.06
62.22
Qwen3Guard-8B-Strict
Scale=8B+
2026.06
61.5
Qwen3Guard-4B-Strict
Scale=4∼7B
2026.06
61.25
Qwen3Guard-8B-Loose
Scale=8B+
2026.06
52.84
ShieldGemma-9B
Scale=8B+
2026.06
51.2
LlamaGuard4-12B
Scale=8B+
2026.06
50.95
NemoGuard-8B
Scale=8B+
2026.06
49.74
ShieldGemma-27B
Scale=8B+
2026.06
47.19
Feedback
Search any
task
Search any
task