SageMaker AI introduces serverless training sessions
What happened
From the reports
AWS 宣布 SageMaker AI 无服务器模型定制中的训练会话正式可用，用户可自行编写并协调训练循环，包括自定义损失函数和奖励信号，GPU 分配、分布式执行、故障恢复与 checkpoint 存储由平台负责。
From AWS Machine Learning Blog
Timeline
Follow the reports to see the event from each side.
Oct 23
- AWS Machine Learning BlogAnnouncing new ways to customize models with SageMaker AI, without managing GPUs
Amazon SageMaker AI serverless model customization now includes training sessions, generally available today. You write and coordinate your own post-training loop (custom loss functions, reward signals, or multi-turn reinforcement learning) while SageMaker AI handles GPU allocation, distributed execution, fault recovery, and checkpoint storage.
Heat of this event
Not enough continuous observations yet to draw a trend.