AWS sagemaker medium security documentation change
Summary
Content replaced with a redirect notice pointing to updated Nova 1.0/2.0 user guides. Removed detailed documentation about Nova model customization workflows, security notes about KMS key usage, and prerequisite references.
Security assessment
The change removes explicit security guidance about KMS key usage requirements (including encryption grants and required permissions for API callers). While this appears to be a relocation rather than a deletion, the removal of these specific security-related operational details from this documentation page could lead to misconfiguration risks if users don't follow the new external links.
Diff
diff --git a/sagemaker/latest/dg/nova-model.md b/sagemaker/latest/dg/nova-model.md index 7859b3351..0b84b040b 100644 --- a//sagemaker/latest/dg/nova-model.md +++ b//sagemaker/latest/dg/nova-model.md @@ -7,41 +7 @@ -You can customize [Amazon Nova models](https://docs.aws.amazon.com//nova/latest/userguide/what-is-nova.html), including the enhanced Nova 2.0 models, through [recipes](./nova-model-recipes.html#nova-model-get-recipes) and train them on SageMaker AI. These recipes support techniques such as supervised fine-tuning (SFT), Direct Preference Optimization (DPO), and Reinforcement Fine-Tuning (RFT), with both full-rank and low-rank adaptation (LoRA) options. - -The end-to-end customization workflow involves stages like model training, model evaluation, and deployment for inference. This model customization approach on SageMaker AI provides greater flexibility and control to fine-tune its supported Amazon Nova models, optimize hyperparameters with precision, and implement techniques such as LoRA parameter-efficient fine-tuning (PEFT), full-rank SFT, DPO, RFT, Continued Pre-Training (CPT), Proximal Policy Optimization (PPO), etc. - -SageMaker AI offers two environments for customizing Amazon Nova models. - - * [**Amazon SageMaker training jobs**](https://docs.aws.amazon.com//sagemaker/latest/dg/how-it-works-training.html) provides a fully managed environment for customizing Nova models where you don't need to create or maintain any clusters. The service automatically handles all infrastructure provisioning, scaling, and resource management, allowing you to focus solely on configuring your training parameters and submitting your job. You can customize Nova models on SageMaker training jobs with techniques like Parameter Efficient Fine-tuning (PEFT), Full rank fine tuning, Direct Preference Optimization (DPO), and Reinforcement Fine-Tuning (RFT). For more information, see [Amazon Nova customization on SageMaker training jobs](./nova-model-training-job.html). - -###### Note - -If you provide a KMS key to your Nova model customization training job for encryption in the Amazon-owned output S3 bucket: - - * You must provide the same KMS key when calling subsequent [iterative training jobs](https://docs.aws.amazon.com//sagemaker/latest/dg/nova-iterative-training.html), or when calling the [CreateCustomModel](https://docs.aws.amazon.com//bedrock/latest/APIReference/API_CreateCustomModel.html#bedrock-CreateCustomModel-request-modelKmsKeyArn) API leveraging the encrypted model. - - * The identity calling the `CreateTrainingJob` API (rather than the execution role) must have permissions to `CreateGrant`, `RetireGrant`, `Encrypt`, and `GenerateDataKey` as defined in KMS key policy. - - * [**Amazon SageMaker HyperPod**](https://docs.aws.amazon.com//sagemaker/latest/dg/sagemaker-hyperpod.html) offers a specialized environment to train Nova models by requiring you to create and manage SageMaker HyperPod EKS clusters with restricted instance groups (RIGs). This environment gives you flexibility in configuring your training environment with specialized GPU instances and integrated Amazon FSx for Lustre storage, making it particularly well-suited for advanced distributed training scenarios and ongoing model development. For more information, see [Amazon Nova customization on Amazon SageMaker HyperPod](./nova-hp.html). - - - - -###### In this chapter - - * [General prerequisites](./nova-model-general-prerequisites.html) - - * [Amazon Nova recipes](./nova-model-recipes.html) - - * [Amazon Nova customization on SageMaker training jobs](./nova-model-training-job.html) - - * [Amazon Nova customization on Amazon SageMaker HyperPod](./nova-hp.html) - - * [Iterative Training](./nova-iterative-training.html) - - * [Amazon Bedrock inference](./nova-model-bedrock-inference.html) - - * [Limitations of customizing Amazon Nova models](./nova-model-limitations.html) - - * [Amazon Nova customization FAQs](./nova-model-faqs.html) - - - +This topic has moved. For the latest information, see Customizing Amazon Nova models on Amazon SageMaker AI in the [Amazon Nova 1.0 user guide](https://docs.aws.amazon.com//nova/latest/userguide/nova-model.html) or the [Amazon Nova 2.0 user guide](https://docs.aws.amazon.com//nova/latest/nova2-userguide/nova-model.html). @@ -57 +17 @@ Customizing models -General prerequisites +Amazon Nova recipes