---
format: "aidr-story-markdown/v1"
id: "189563f1d6f7d0a3fade791a901e4649945d79a7b51ee71de60ed7235117b024"
canonical_url: "https://aidr.today/189563f1?lang=en"
title: "SageMaker Inference now supports Multi-Instance GPU to maximize GPU utilization for model endpoints"
lang: "en"
requested_lang: "en"
available_langs: ["en","vi"]
translation_fallback: null
fallback_fields: []
published_at: "2026-11-06T18:01:00.000Z"
category: "Tools"
topics: ["amazon","inference","devtools"]
source_urls: ["https://aws.amazon.com/blogs/machine-learning/sagemaker-inference-now-supports-multi-instance-gpu-to-maximize-gpu-utilization-for-model-endpoints/"]
summary: "Amazon SageMaker Inference now supports GPU partitioning with NVIDIA Multi-Instance GPU. This feature divides one GPU into hardware-isolated partitions to run a different model in each or use a right-sized fraction of a newer accelerator."
---

# SageMaker Inference now supports Multi\-Instance GPU to maximize GPU utilization for model endpoints

> [Open the canonical story](<https://aidr.today/189563f1?lang=en>)

**Published:** 2026-11-06T18:01:00.000Z
**Category:** Tools
**Topics:** amazon, inference, devtools

## Summary

Amazon SageMaker Inference now supports GPU partitioning with NVIDIA Multi\-Instance GPU\. This feature divides one GPU into hardware\-isolated partitions to run a different model in each or use a right\-sized fraction of a newer accelerator\.

## Sources

- [Story source](<https://aws.amazon.com/blogs/machine-learning/sagemaker-inference-now-supports-multi-instance-gpu-to-maximize-gpu-utilization-for-model-endpoints/>)