扩展定律 / 规模法则Scaling Laws
扩展定律 / 规模法则是什么?
表明 AI 模型性能随算力、数据与参数规模扩大而可预测提升的经验规律。2025-2026 年，焦点从预训练扩展拓展到面向推理模型的推理时（测试时）算力扩展。
What is Scaling Laws?
Empirical relationships showing that AI model performance improves predictably as compute, data, and parameters scale up. In 2025-2026 the focus broadened from pre-training scaling to inference-time (test-time) compute scaling for reasoning models.
天际投资视角 · Agent 与企业智能
Agent 正从演示走向生产环境。价值向两端集中——掌握业务工作流的应用层,和按 token 计价的基础设施;中间的通用薄壳会被挤掉。