NousResearch released Arete-27B, a post-trained variant of Alibaba's Qwen3.8-27B designed to reinforce Western values and address observed refusals and CCP-framed responses in the base model. The GGUF quantized version is available on Hugging Face via bartowski, enabling local deployment. This work demonstrates how open-source models allow labs to reshape model alignment according to diverse user preferences.