Amazon SageMaker Launches G7e Instances to Boost Generative AI Performance

Published on April 20, 2026

Amazon has expanded its cloud capabilities with the introduction of G7e instances within SageMaker AI. This addition features NVIDIA RTX PRO 6000 Blackwell Server Edition GPUs, which allow for enhanced generative AI inference. Users can now select configurations with varying numbers of GPUs, catering to different computational needs.

The launch addresses a growing demand for robust AI infrastructure. Organizations increasingly seek cost-effective solutions for deploying powerful open source foundation models. The G7e instances promise to deliver the necessary performance without excessive expenditure, effectively lowering barriers to entry.

Key features include support for single-node setups as well as configurations with up to eight GPUs. Each GPU is equipped with 96 GB of GDDR7 memory, ensuring ample resources for demanding applications. Models like GPT-OSS-120B and Nemotron-3-Super-120B-A12B can now be implemented more efficiently.

This rollout could significantly accelerate the development of AI applications across various sectors. Businesses now have access to scalable, high-performance options that can streamline workflows and enhance productivity. As a result, the landscape for generative AI is poised for transformation, paving the way for innovation.

Related News