STORY · PRODUKTER_
Mistral launches Shieldstral – 3B multimodal safety classifier for content moderation
Mistral AI has launched Shieldstral, an open 3B model for safety and content moderation that accepts natural language guidelines during inference. The model performs competitively with larger models, runs on a single 16GB GPU, and is released under the Apache 2.0 license.
WHY IT MATTERS
Shieldstral solves a practical problem for AI products: it enables flexible, policy-adaptive moderation without retraining, making it easier to adapt safety to different user communities and domains.
SOURCES
MACHINE-GENERATED SUMMARY This summary is written by machine from the sources below. We sort and explain — but we are a way into the field, not the final word. Check the source when something matters to you.