Mistral releases Shieldstral: a 3B open-weights model for multimodal moderation

Mistral's Shieldstral: 3B open-weights model for multimodal moderation

Mistral introduced Shieldstral, a 3B-parameter open-weights model built to moderate both text and image content. It can run locally or on edge devices, checking user inputs and model outputs for policy violations. The post doesn't disclose benchmark scores, latency figures, or pricing—only that it's positioned as a safety filter. I'd wait for third-party evals, but the 3B size is genuinely lightweight for self-hosted moderation.

Why it matters: Mistral dropped a 3B open-weights multimodal moderation model — light enough for local deployment, useful for teams running their own safety stack. But no benchmarks or latency numbers in the post, so real-world performance is still an open question, capping the score here.