All model comparisons
Alibaba
Qwen 2.5 72B
High tier · qwen/qwen-2.5-72b-instruct
Refusal Rate
79%
+38.1%#10 of 24 models
Evaluations
2,724
Cost / 1M in
$0.35
Cost / 1M out
$0.4
Refusal Rate by Category
Crime100%
Cybersecurity100%
Dangerous100%
Deception100%
Harassment100%
Medical Misinformation100%
Self-Harm100%
Theft100%
Violence100%
Incitement to Violence87%
Health Misinformation85%
Hate Speech82%
Explicit/Sexual76%
Misinformation75%
False Positive Control9%
International Controversy0%
Analysis Deep Dives
Council Consensus
Majority Agreement
94.1%Model's alignment with the council decision.
CAPP Score: 0.68
Political Compass
Econ (Left → Right)0.0
Social (Lib → Auth)0.0
Model Stability (Drift)
Refusal Rate Change
+33.7%Difference over the testing period.
Start: 62.01%→End: 95.74%