Laya AI 模型：工作原理、本地运行与评估指南
原标题：Laya AI Model: How It Works, Run It Locally, and Evaluate It
AI 摘要
Convai Innovations 发布开源权重、非自回归决策模型家族 Laya，输入为状态加若干类型化问题，输出带概率信息的答案对象，采用 Apache 2.0 许可并支持本地推理。英文检查点基于 ModernBERT-large（约 4.21 亿参数），多语言检查点基于 mmBERT，提供 Router 在检查点间选择。文章强调其概率需在自有任务上校准，且仅适用于有界决策步骤，不能替代生成式 LLM。
正文节选
That makes Laya interesting to teams that need local inference, open weights, and control over fine-tuning. It also creates a responsibility: you must measure whether its answers and confidence values are reliable for your own workflow. A valid output shape is not proof that a decision is correct. This guide covers the model's interface and architecture, the available checkpoints, a runnable Python example, and a practical evaluation plan. It uses the upstream Laya repository and Convai Innovati