AWS First Steps
LanguageEnglish
New feature

Amazon SageMaker AI now supports custom sequence lengths for serverless model customization jobs

Amazon SageMaker AI now allows users to configure sequence lengths in serverless model customization jobs, enabling control up to 16K tokens to better match training data characteristics.

Amazon SageMaker AI now allows configuring sequence lengths for serverless model customization jobs. This update enables fine-tuning popular model families like gpt-oss, Gemma, Llama, Nemotron, and Qwen. Previously fixed, sequence lengths can now be set via the dataset_max_len hyperparameter for SFT and DPO jobs, and max_prompt_length and max_response_length for RFT jobs, supporting everything from short Q&A pairs to multi-page documents. Most models support up to a maximum of 16K tokens. Serverless model customization is available in US East (N. Virginia), US West (Oregon), Asia Pacific (Tokyo), and Europe (Ireland). See SageMaker AI documentation for details.

Why it matters

This change impacts users fine-tuning LLMs, especially those working with long text data or document processing requirements.

Regions: N. Virginia, Oregon, Tokyo, Ireland

Read the original AWS announcement

Related articles and terms

The Layered Architecture of AWS AI/ML Services - Flexibility Through the Three Tiers of SageMaker, Bedrock, and API ServicesThis article organizes AWS AI/ML services into three layers - SageMaker (full control), Bedrock (managed generative AI), and Rekognition/Comprehend/etc. (API-based) - and explains AWS's flexibility through comparisons with GCP Vertex AI and Azure OpenAI Service, including custom silicon integration.Amazon SageMakerA fully managed platform for building, training, and deploying machine learning models end-to-end, significantly boosting productivity for data scientists and ML engineersBuilding an ML Platform with Amazon SageMaker - From Model Development to DeploymentFrom development in Studio to managed spot training, MLOps with Pipelines, and data drift detection with Model Monitor, this article covers how to integrate the entire ML lifecycle.Generative AI Platform - Building an Enterprise AI Foundation with Amazon BedrockLearn how to build generative AI applications using Amazon Bedrock. This guide covers foundation model selection, RAG pattern implementation, safety guardrails, and SageMaker integration for designing enterprise-grade AI infrastructure.GPU-Based Machine Learning Training with AWS Batch - Cost-Efficient Large-Scale TrainingRun GPU training with your existing Docker containers, and cut costs by up to 90% using Spot Instances and checkpointing. Includes guidance on when to choose Batch over SageMaker.