Amazon SageMaker AI in AWS European Sovereign Cloud - AWS European Sovereign Cloud User Guide
How Amazon SageMaker AI differsDocumentation
Services or capabilities described in AWS documentation might vary by Region. To see the differences applicable to the AWS European Sovereign Cloud Region, see the AWS European Sovereign Cloud User Guide.

Amazon SageMaker AI in AWS European Sovereign Cloud

This topic describes the functionality of Amazon SageMaker AI in the AWS European Sovereign Cloud Region.

Amazon SageMaker AI is a fully managed platform that provides every developer and data scientist with the ability to build, train, and deploy machine learning (ML) models quickly. SageMaker AI removes the heavy lifting from each step of the machine learning process to make it easier to develop high quality models.

How Amazon SageMaker AI differs

The following differences apply to Amazon SageMaker AI:

  • Only the following features are available:

    • Training Jobs - Fast file input mode and IFDP are not supported.

    • Processing Jobs - Processing jobs that use SageMaker Spark Container Images are not supported.

    • Real-time Inference - The following features are not supported:

      • Auto scaling

      • Asynchronous Inference

      • Serverless Inference

      • Inference Components

    • Studio

      • Authentication using AWS Identity and Access Management is supported; authentication using IAM Identity Center is not supported.

      • Scheduling a notebook job is not supported.

      • Only VPC option is available, and public internet is not available.

    • AWS Deep Learning Containers (DLC) - Images are published to a private Amazon ECR registry in AWS European Sovereign Cloud. Use the following information to authenticate to the registry and construct image URIs:

      • Registry account - 204133271717 hosts the DLC images (for example, base and vLLM).

      • ECR login - authenticate to the registry:

        aws ecr get-login-password --region eusc-de-east-1 | docker login --username AWS --password-stdin 204133271717.dkr.ecr.eusc-de-east-1.amazonaws.eu

      • Image URI format:

        204133271717.dkr.ecr.eusc-de-east-1.amazonaws.eu/<REPOSITORY>:<TAG>

        For example, the base repository with the tag runtime-cu130-amzn2023:

        204133271717.dkr.ecr.eusc-de-east-1.amazonaws.eu/base:runtime-cu130-amzn2023

      • Available repositories - sagemaker-scikit-learn and sagemaker-xgboost are not available.

      • Available tags - only mutable tags are available in this Region. A mutable tag can be repointed to a newer image when a version is patched or updated, so pulling the same tag at a later date may return different image content. Immutable tags, which pin to a fixed, unchanging build, are generally not published, apart from a few exceptions required by specific consumers (for example, the vLLM -v1.x SageMaker tags). To illustrate the mutable-tag scheme:

        • Ubuntu-based vLLM - available tags: 0.25.1-gpu-py312-cu130-ubuntu22.04-ec2, 0.25.1-gpu-py312-ec2. Immutable tags such as 0.25.1-gpu-py312-cu130-ubuntu22.04-ec2-v1.2-2026-07-20-21-30-05 and 0.25-gpu-py312-cu130-ubuntu22.04-ec2-v1 are not available.

        • AL2023-based vLLM - available tags: server-cuda-v2.1, server-cuda-v2, server-cuda. Immutable tags such as server-cuda-v2.1.3 are not available.

        • For the full, current list of tags in a repository, use the standard Amazon ECR CLI commands (aws ecr list-images / aws ecr describe-images) against the eusc-de-east-1 Region.

      • Not all framework versions are actively patched. Versions follow the standard DLC Support Policy - use a currently supported version to ensure you continue receiving security patches.

    • Model Cards

    • Model Registry

    • Pipelines

    • Python SDK

    • JupyterLab

    • Search - Search does not return results for SageMaker AI features that are not available.

    • Metrics - Public endpoints are not supported; only internal endpoints are supported.

    • Distribution

    • Images

    • Role Manager

Documentation