Introducing Arete-27B: Qwen3.8-27B, post-trained to reinforce Western values.
This is a dev post classified by AGK Intelligence as Other (a launch), kept by the Dev Radar because it carries real work, not commentary.
Introducing Arete-27B: Qwen3.8-27B, post-trained to reinforce Western values. Chinese open models are currently the strongest available, but they ship with state-aligned behavior in the weights. We built an 803-prompt eval across seven value categories, including freedom of expression, democracy, and privacy, and measured how often a model adopts China’s party line. After fine-tuning on roughly 53 million tokens, Arete-27B's propaganda rate fell from 22% to 1%, in line with the Western models we tested. Its Hermes Bench score shows no significant performance tradeoffs. The values in an open
Posted by Nous Research (296.2k followers) 4 h ago · 4.9k likes · 198.8k views · view the original post on X. Kept by the Dev Radar as Other.
More dev work like this
- 10/10 comms — @hosseeb
- Want to see the difference in answers between a Qwen model as is, and with its… — @tonbistudio
- ⭐️PS3 Game Cheats updated by mohasi — @aldostools
- the tech world needs more of this x100 — @theaaron
- Let an AI agent help run your smart home — without putting it in the real-time control… — @DanKornas
- 英文视频想配字幕，最难缠的往往不是翻译本身，而是网站、模型、导出格式来回倒腾。 — @bkdgiffug
- 好家伙！现在做 PPT，排版都不用自己折腾了！ — @bkdgiffug
- Vendor submissions → AI risk tiers → human review for higher-risk cases. — @Box
Every post is read and classified by AGK Intelligence: what it is, which market it belongs to, and whether the link is a real tool. 32k posts from 5k X accounts over the last 21 days, 9.5k tools, 12 markets. New posts are collected several times an hour and the full ranking is recomputed every hour — last update 2026-11-07 02:41 UTC. Full method.