We introduce A.X K1, a 519B-parameter Mixture-of-Experts (MoE) language model trained from scratch. Our design leverages scaling laws to optimize training configurations and vocabulary size under fixed computational budgets. A.X K1 is pre-trained on a corpus of approximately 10T tokens, curated by a multi-stage data processing pipeline. Designed to bridge the gap between reasoning capability and inference efficiency, A.X K1 supports explicitly controllable reasoning to facilitate scalable deployment across diverse real-world scenarios. We propose a simple yet effective Think-Fusion training recipe, enabling user-controlled switching between thinking and non-thinking modes within a single unified model. Extensive evaluations demonstrate that A.X K1 achieves performance competitive with leading open-source models, while establishing a distinctive advantage in Korean-language benchmarks.

翻译：我们介绍了 A.X K1，这是一个从头开始训练的、拥有 5190 亿参数的专家混合模型。我们的设计利用缩放定律，在固定的计算预算下优化了训练配置和词汇表大小。A.X K1 在一个约 10 万亿词元的语料库上进行预训练，该语料库由一个多阶段数据处理流程精心构建。该模型旨在弥合推理能力与推理效率之间的差距，支持显式可控的推理，以促进在不同现实场景中的可扩展部署。我们提出了一种简单而有效的 Think-Fusion 训练方法，使得用户可以在一个统一的模型内控制式地切换思考模式与非思考模式。广泛的评估表明，A.X K1 的性能与领先的开源模型相当，同时在韩语基准测试中建立了独特的优势。