Skip to main content

Shieldstral 1.0 (3B)

Mistral AIOpen WeightApache 2.0 · Commercial OK

Description

Shieldstral is a 3B open-weight multimodal safety classifier from Mistral. It frames content moderation as policy-adaptive yes/no question answering: plain-language policies are supplied at inference time, and the model returns a calibrated safety score from a single forward pass for text, image, or text+image content. Released under Apache 2.0. Aggregate self-reported average F1: text-safety 0.849, multimodal 0.838 (HF mistralai/Shieldstral-1.0-3B / arXiv 2607.25857); per-bench WildGuard/HarmBench/ToxicChat/BeaverTails/Aegis/VLGuard ids are not in the catalog.

Release Date
2026-08-04
Parameters
3.0B
Context Length
Modalities

Capability Radar

90
general
0
coding
0
reasoning
0
scienceest.
0
agents
0
multimodal

Science is estimated from LLM Stats science scores or reasoning when no dedicated science benchmarks are available.

Rankings

No ranking data available

Benchmark Scores (LLM Stats)

(LLM Stats (zeroeval))

Safety

XSTest94.6%SR

AA Evaluation Indices

(Artificial Analysis)

No AA evaluation data available

LLM Stats Category Scores

(LLM Stats (zeroeval))
Safety
90
General
90

Pricing

No pricing data available

Speed

No speed data available

Provider Price Ranking

No provider data available

External Sources