Shieldstral 1.0 3B
A 3B open-weight multimodal safety classifier that applies plain-language policies to prompts, responses, refusals, toxicity, and images.
0 comments · 0 votes
Sign in to join the discussion →
No comments yet. Start the discussion.
Why this model scores 20.1
Shieldstral is highly deployable because its Apache 2.0 weights fit a single 16 GB GPU, but it is a narrow moderation classifier with no demonstrated autonomous agency or general-purpose capability.
News tied to Shieldstral 1.0 3B
The model score of 20.1 rates this model's risk profile. The overall Doom Index of 66.8 measures the complete temporally weighted evidence record. These values answer different questions.
Each article's current Doom Index contribution is divided equally among the exact models named on that article. This prevents multi-model evidence from being claimed in full on several model pages. Model risk scores use a bounded temporal offset around their technical profile, but never feed back into the overall index.
Mistral releases open Shieldstral safety classifier
Mistral released Shieldstral 1.0 3B under Apache 2.0 as a policy-adaptive text and image safety classifier, with held-out benchmark results and operation on a single 16 GB GPU.
- Full item contribution
- -0.07
- Shieldstral 1.0 3B equal share
- -0.07
Model score history
-
R1
Doom Score 20.1
Mistral's dated release, official model card, and weight repository establish Shieldstral 1.0 3B as an open, policy-adaptive safety classifier.
13 Aug 2026