Introducing Shieldstral.
Shieldstral introduces a 3B open-weights multimodal safety classifier that outperforms models up to 7x its size.
Get smart on it
Shieldstral is a small open-weights multimodal safety classifier that evaluates whether text and images meet specified safety policies by framing content moderation as a question-answering task. Rather than baking fixed harm categories into a model that requires retraining for new contexts, Shieldstral accepts plain-language policies at inference time, allowing a single model to adapt to different applications and audiences without modification. The model performs comparably to much larger competitors while running efficiently on standard hardware, and it produces calibrated safety scores that can be thresholded rather than relying on discrete labels. This approach matters because different products, domains, and audiences require different safety definitions, and existing guardrail models force all deployments through one frozen set of categories.
Related news
Pokee AI Releases Pokee-Isaac 28B: A 10M-Token Context Agentic Model Built to Run Inside the Customer Boundary
Long-horizon agents accumulate context faster than they resolve tasks. Every tool output, observation, and intermediate reasoning step stays in the window, and the two capabilities that matter — holding that context and staying coherent across it — have so far been available almost exclusively from cloud endpoints. That excludes regulated industries, public-sector institutions, and on-device applications, where the data is not permitted to leave the boundary at all. Pokee AI released Pokee-Isaa
Mistral AI Releases Shieldstral 1.0 3B: An Open-Weights Policy-Adaptive Multimodal Safety Classifier Matching Models 7× Its Size
Mistral AI has released Shieldstral 1.0 3B, an open-weights, policy-adaptive multimodal safety classifier that treats content moderation as a single yes/no question rather than a fixed taxonomy of harm categories. Most guardrail models bake their category list into the weights, so re-targeting one to a new deployment context means retraining — and the same content can be acceptable on a cybersecurity research tool while being harmful on a mental-health platform. Shieldstral inverts that: operat
