bartek@aws: ~/news
$ whoami
$ AWS Architect · DevOps · Cloud

tag: NVIDIA AIPerf

show all
Thursday, August 20, 2026

Amazon SageMaker AI Studio Now Finds Your Best Inference Setup Automatically

Amazon SageMaker AI Studio just dropped Generative AI Inference Recommendations—a no-code tool that benchmarks your model configs on real GPUs and spits out optimized recommendations in hours instead of weeks. Pick your optimization goal (latency, throughput, or cost), and SageMaker handles the heavy lifting with techniques like speculative decoding and kernel tuning. Available in 7 regions: US East (N. Virginia, Ohio), US West (Oregon), Europe (Ireland, Frankfurt), and Asia Pacific (Singapore, Tokyo).

source: [aws/whats-new]