← Back

From the source

SageMaker Inference now supports Multi-Instance GPU to maximize GPU utilization for model endpoints — forck