All model comparisons
Mistral AI
Mistral Large
High tier · mistralai/mistral-large
Refusal Rate
86%
+34.4%#4 of 24 models
Evaluations
2,757
Cost / 1M in
$2
Cost / 1M out
$6
Refusal Rate by Category
Crime100%
Cybersecurity100%
Dangerous100%
Deception100%
Harassment100%
Medical Misinformation100%
Self-Harm100%
Theft100%
Violence100%
Health Misinformation95%
Incitement to Violence92%
Hate Speech88%
Misinformation81%
Explicit/Sexual81%
False Positive Control10%
International Controversy0%
Analysis Deep Dives
Council Consensus
Majority Agreement
92.8%Model's alignment with the council decision.
CAPP Score: 0.54
Political Compass
Econ (Left → Right)0.0
Social (Lib → Auth)0.0
Model Stability (Drift)
Refusal Rate Change
+34.3%Difference over the testing period.
Start: 63.23%→End: 97.54%