Deploying Kimi K3 on Amazon SageMaker HyperPod and Amazon EKS
This post walks through deploying Kimi K3 on AWS using two approaches: Amazon SageMaker HyperPod, and Amazon Elastic Kubernetes Service (Amazon EKS) cluster.
Follow AWS Machine Learning Blog to make it a durable For You signal.
AWS published a guide on July 30, 2026 detailing how to deploy Kimi K3, a 2.8-trillion-parameter Mixture of Experts open-weight model released by Moonshot AI, using Amazon SageMaker HyperPod or Amazon Elastic Kubernetes Service (Amazon EKS). The guide describes the model's specifications, required p6-b300 GPU instances, capacity reservation methods, deployment configurations, and invocation examples, without offering performance benchmarks or independent verification of the model's claims.