Category: Compute

Learn how to deploy one large language model across two GPU nodes on Amazon EKS using the multi-node Ray LLM Deep Learning Containers, with separate prefill and decode nodes behind one OpenAI-compatible endpoint. The pre-built container ships the full inference stack, so you configure the deployment instead of assembling it.

Learn how to scale Mixture-of-Experts (MoE) reinforcement learning on Amazon EKS using Elastic Fabric Adapter (EFA) and DeepEP. This post presents an architecture that combines Amazon EKS, EFA, and Amazon S3 and increased aggregate reinforcement learning rollout throughput by 40% for large-scale RLHF and GRPO training.

Integrate NVIDIA Resiliency Extension (NVRx) into PyTorch FSDP training on Amazon EKS to overlap checkpoint I/O with training and recover from GPU faults in seconds. This post covers async checkpointing, in-process restart, and ft_launcher in-job restart, with H100 benchmarks at 2 to 8 nodes showing 99%+ training efficiency and second-scale recovery.

Benchmark two 30B Mixture-of-Experts models, Qwen3-Coder-30B and NVIDIA Nemotron-3-Nano-30B, across G5, G6, G6e, and G7 GPU instances on Amazon SageMaker AI. Compare throughput, latency, and cost-per-token, and see how G7’s NVIDIA Blackwell GPUs deliver measurable price-performance gains for real-time LLM inference.

Deploy a customer-operated LiteLLM gateway on Amazon ECS with AWS Fargate, connect it to an OpenAI model on Amazon Bedrock, and configure Codex to route requests through the gateway’s Responses API with scoped identities, budgets, rate limits, and telemetry. We also compare direct IAM Identity Center access and a managed Portkey deployment.

Serving automatic speech recognition (ASR) models at scale is costly when each request uses only a fraction of a GPU. Learn how NVIDIA CUDA Multi-Process Service (MPS) with NVIDIA Triton Inference Server on Amazon EC2 GPU instances cuts GPU infrastructure by 75% while holding sub-second latency at 92.1 requests per second per GPU.

ONESTRUCTION, with technical advisory from the AWS Generative AI Innovation Center, built Ishigaki-IDS, a foundation model specialized for construction and BIM workflows. This architectural case study shows how they combined synthetic data, a three-stage training pipeline, and verifiable rewards on Amazon EC2 to build a domain model in a data-scarce field.

The Amazon SageMaker AI Spaces add-on for Amazon EKS runs managed JupyterLab and Code Editor environments on the cluster your ML team already operates. This post shows how to install and configure the add-on, connect from the browser and from VS Code over SSH-over-SSM, and move your team to OpenID Connect sign-in with Amazon Cognito.

In this post, we’ll explore how Mobileye deployed an AI support agentic solution on Amazon Bedrock AgentCore – from the support bottleneck that sparked the idea, through the proof of concept that validated it, to the hybrid architecture that bridges on-premises systems with AWS cloud services. This approach is relevant for enterprises struggling to scale AI Agents while maintaining enterprise grade governance and security standards.

This post walks through deploying Kimi K3 on AWS using two approaches: Amazon SageMaker HyperPod, and  Amazon Elastic Kubernetes Service (Amazon EKS) cluster.