Mistral 发布 3B 多模态安全分类器 Shieldstral
Shieldstral 用自然语言政策替代固定分类，一个 3B 模型同时审核文本和图像，性能匹敌 7 倍大小模型。
Mistral 推出开源 3B 多模态安全分类器 Shieldstral。它把内容审核建模为策略自适应问答任务，可在推理时接受自然语言政策，无需重新训练即可统一审核文本和图像。在多项基准上，其表现不输于 7 倍规模的模型，且仅需一块 16GB GPU 即可运行。
正文摘录
Solutions Introducing Shieldstral. August 4, 2026 By Mistral [Back to Blog](https://mistral.ai/news/)5 min read Share this post Copy url to clipboard Copied   Thinking Summary Shieldstral introduces a 3B open-weights multimodal safety classifier that outperforms models up to 7x its size by framing content moderation as a policy-adaptive question-answering task. Unlike traditional guardrail models, it accepts plain-language policies at inference time, unifying text and image safety evaluation without retraining. Released under Apache 2.0, it delivers calibrated safety scores across diverse benchm…