Back to News

Nous Research unveils Arete-27B, a post-trained Qwen3.8-27B model reinforcing Western values

#nous research#arete-27b#qwen#post-training

Nous Research announced Arete-27B, a post-trained version of Alibaba's Qwen3.8-27B model, designed to reinforce Western values. The company claims the underlying Chinese model's bias layer is shallow, justifying the additional post-training.

Coverage timeline

  1. Techmeme
    Nous Research unveils Arete-27B, a post-trained Qwen3.8-27B model that reinforces Western values, saying the underlying Chinese model's bias layer is shallow (Nous Research)

    Nous Research : Nous Research unveils Arete-27B, a post-trained Qwen3.8-27B model that reinforces Western values, saying the underlying Chinese model's bias layer is shallow — We post-trained over Alibaba's Qwen3.8-27B to reinforce Western values on a Chinese-trained model. We call this new model Arete-27B.

Related stories

  • MIT: GPT-6 Astra edits Qwen weights, boosting MATH-500 from 7.9% to 48.1%

    MIT CSAIL researchers demonstrated that GPT-6 Astra can directly read and modify the weights of Qwen3.5-0.8B, improving its MATH-500 score from 7.9% to 48.1% by altering only 0.016% of parameters, without training or gradient descent. The team named this capability Weight-Space Programming. However, spaCy author Matthew Honnibal publicly challenged the paper, arguing that the experimental conditions across the three claimed abilities differ significantly and that the presentation is misleading.

    6 reports · 5 sources2949 points · 892 comments

  • BAAI AREX-2: 27B model beats larger rivals on ML, algorithm benchmarks

    The AREX research group, involving BAAI, USTC, and Renmin University, released AREX-2, a 27B-parameter model built on Qwen3.8-27B and trained on long-horizon reflection data for machine learning and algorithm programming. It outperforms larger models on MLE-bench Lite and Frontier-CS, and sets open-weight records on BrowseComp and HLE, suggesting that long-horizon reflection training can foster cross-domain self-improvement capabilities.

    1 reports · 1 sources

  • Chinese AI startups HyperVolve and Apexin announce funding and model releases

    Chinese AI startup HyperVolve announced tens of millions RMB in angel funding led by Lanchi Ventures, with participation from SenseTime Guoxiang Capital, Lenovo Star, Inno Fund, and Tsinghua alumni seed fund, to build research foundation models using recursive self-improvement. Separately, Apexin, founded by 27-year-old Tsinghua assistant professor Chen Yongchao, released Apex-35B-V1.0, a 35B model claiming SOTA results on 21 benchmarks against comparable models.

    2 reports · 1 sources

  • Microsoft Announces Decision-1, a 9B Qwen-Based Model

    Microsoft has announced Decision-1, a 9-billion-parameter language model built on the Qwen architecture. The announcement details the model's launch and its key features, though specific technical specifications and availability were not disclosed in the provided sources.

    1 reports · 1 sources71 points · 31 comments

News updates every hour. Subscribe in your RSS reader to get new stories automatically.

Subscribe