Storia in 5 fonti

Introducing Shieldstral. | Mistral AI

Shieldstral introduces a 3B open-weights multimodal safety classifier that outperforms models up to 7x its size.

Raccontata da

Confronto fonti

5 prospettive sulla stessa storia

AI · summaries

mistral.aiStai leggendo1 mese fa

Introducing Shieldstral. | Mistral AI

Mistral AI released Shieldstral, a 3B policy-adaptive safety classifier matching models 7x larger without retraining when policies change. One deployment adapts to any moderation policy across domains, reducing governance overhead for foundation model builders.

originale

Timeline cronologica

martedì 4 agosto 2026·mistral.ai
Introducing Shieldstral. | Mistral AI
Shieldstral introduces a 3B open-weights multimodal safety classifier that outperforms models up to 7x its size.
mercoledì 5 agosto 2026·dev.to
Shieldstral Introduces Policy-Adaptive Multimodal Safety Classification in a 3B Model
Shieldstral is a 3B-parameter policy-adaptive multimodal safety classifier designed to assess text...

the-decoder.com

1 mese fa

Mistral's open model Shieldstral matches much larger safety models at a fraction of the size

Mistral's 3B Shieldstral ties 20B GPT-OSS-Safeguard on safety benchmarks (F1 84.9%) via single-token answers. Operators define custom filtering rules at runtime per-app without retraining, reducing classifier overhead and costly false positives on production.

Leggi questa versione → originale

siliconangle.com1 mese fa

Mistral introduces Shieldstral to provide lightweight policy-aware moderation for AI models - SiliconANGLE

Mistral AI releases Shieldstral (3B params), outperforming models 7x larger on safety (84.9% text, 83.8% image), deployable on single 16GB GPU. Natural language policy configuration at inference enables compliance guardrails without model retraining, addressing governance friction in enterprise AI deployments.

Leggi questa versione → originale

dev.to1 mese fa

Shieldstral Introduces Policy-Adaptive Multimodal Safety Classification in a 3B Model

Shieldstral is a 3B-parameter policy-adaptive multimodal safety classifier designed to assess text...

Leggi questa versione → originale

heise.de1 mese fa

Mistral veröffentlicht Modell mit neuer Technik für Sicherheitsklassifikationen

Mistrals Shieldstral klassifiziert die Sicherheit von KI-Inhalten mithilfe einer neuen Herangehensweise. Sie soll sich flexibel an veränderte Kontexte anpassen.

Leggi questa versione → originale

mercoledì 5 agosto 2026·

siliconangle.com

Mistral introduces Shieldstral to provide lightweight policy-aware moderation for AI models - SiliconANGLE

mercoledì 5 agosto 2026·

the-decoder.com

Mistral's open model Shieldstral matches much larger safety models at a fraction of the size

Mistral's new 3B Shieldstral model checks AI inputs and outputs for safety violations using natural language yes-or-no questions instead of fixed categories. It matches models…

giovedì 6 agosto 2026·

heise.de

Mistral veröffentlicht Modell mit neuer Technik für Sicherheitsklassifikationen

Mistrals Shieldstral klassifiziert die Sicherheit von KI-Inhalten mithilfe einer neuen Herangehensweise. Sie soll sich flexibel an veränderte Kontexte anpassen.