Mistral AI Open-Sources Shieldstral: A Multimodal Safety Guardrail Model
MistralAI · x · 2026-08-05
Mistral AI announced that Shieldstral is now available under the Apache 2.0 license.
The model takes moderation policies as plain-language inputs and returns calibrated scores for both text and images via a single interface. A comprehensive technical report detailing its architecture and usage has also been released.
Related event: Mistral AI Open-Sources Shieldstral, a 3B Multimodal Safety Model(5 posts)→
More from Models
- Ahead of FLUX3 Release, Developers Fear Over-Censorship Could Ruin the Model — cocktailpeanut · 2026-08-05
- Ant Group's Ling-3.0-flash Tops Hugging Face Trending Models — inclusionAI · 2026-08-05
- Real-World Coding Eval: KAT Coder Outperforms Qwen and Ornith in 35B Local MoE Models — Undici77 · 2026-08-05
- Study: GLM-5.2 Nears Frontier Capabilities but Fails to Refuse Dangerous Tasks — RebeccaBellan · 2026-08-05
- 20B Model Maple-Preview Runs at 200+ tokens/s on Mac Mini, Solves IMO Math — tylerbruno05 · 2026-08-05
- SaferAI Report: Open-Weight Models Approach Frontier Capabilities, But Safety Gap Remains — TechCrunch AI · 2026-08-05