Skip to main content

Model Deployment

Deploy models for real-time and batch inference at scale.

Real-time Inference​

  • SageMaker endpoints
  • Auto scaling configuration
  • Multi-model endpoints

Batch Inference​

  • Batch transform jobs
  • Large-scale processing
  • Cost-effective inference

Edge Deployment​

  • SageMaker Edge Manager
  • IoT device deployment
  • Model optimization for edge

A/B Testing​

  • Multi-variant endpoints
  • Traffic splitting
  • Performance comparison