Mistral AI has released Shieldstral 1.0 3B, an open-weights, policy-adaptive multimodal safety classifier that treats content moderation as a single yes/no question rather than a fixed taxonomy of harm categories. Operators write the policy as a plain-language question at inference time, and the model returns a calibrated safety score from a single forward pass. Built on Ministral-3-3B-Base-2512 with a native Pixtral vision encoder and released under Apache 2.0, it reports 84.9% average F1 on text safety — matching GPT-OSS-Safeguard-20B — and 83.8% on multimodal safety, ahead of every baseline Mistral evaluated. Shieldstral-1.0-3B fits in 16GB of VRAM in BF16, runs on a single GPU, and is licensed Apache 2.0 for commercial and non-commercial use. Serving paths include vLLM (≥0.26.0, recommended), llama.cpp via GGUF conversion with Q8_0/Q5_K_M/Q4_K_M quantization, SGLang, and Transformers, with fine-tuning supported through Axolotl. The classifier emits one token, so latency and cost sit far below reasoning-based guards like GPT-OSS-Safeguard-20B. The 16GB footprint puts it within reach of seed-stage AI product teams, while the open license and self-hosting story suit