A company wants to gradually roll out a new recommendation model (Model B) while keeping the existing model (Model A) in production. They want to route 10% of traffic to Model B and 90% to Model A and compare real-time performance metrics. Which SageMaker feature enables this?

Question

Accepted Answer

C. SageMaker Production Variants with traffic weighting. SageMaker endpoint Production Variants allow deploying multiple model variants behind a single endpoint with configurable traffic weights. Setting Model A weight to 0.9 and Model B weight to 0.1 routes traffic proportionally for A/B testing.

A company wants to gradually roll out a new recommendation model (Model B) while keeping the existing model (Model A) in production. They want to route 10% of traffic to Model B and 90% to Model A and compare real-time performance metrics. Which SageMaker feature enables this?

Related Questions

Discussion