1
Mistral's Shieldstral: 3B open-weights model for multimodal moderation
Shieldstral introduces a 3B open-weights multimodal safety classifier that outperforms models up to 7x its size.
Frontier open-weight models you own and control. Customize with Forge, deploy with Studio, run on-prem, your VPC, or our cloud.
Visit mistral.ai → RSS feedOn Hacker News Sep 2023 – Oct 2026
AI & Machine Learning 42Tech Industry & Business 24Software Engineering 7Tools & Productivity 4Web & Frontend 1Security & Privacy 1
Shieldstral introduces a 3B open-weights multimodal safety classifier that outperforms models up to 7x its size.