Mistral AI Unveils Shieldstral: Policy-Adaptive 3B Open-Weights Multimodal Safety Classifier

Shieldstral is a 3B open-weights multimodal safety classifier that outperforms models up to 7x its size by framing content moderation as a policy-adaptive question-answering task. It enables real-time policy adjustments without retraining, offering calibrated safety scores across text and images under Apache 2.0 licensing.

Comments

Read original article