NVIDIA AI Enterprise Adds Support for NVIDIA H200 NVL
Table of Contents
- Introduction
- AI and Data Science Software Catalog
- Infrastructure Software Collection
- NVIDIA AI Enterprise Software Reference Architecture
1. Introduction
NVIDIA AI Enterprise is a cloud-native software platform designed for developing and deploying production-grade AI solutions. The latest release of the software includes support for the new NVIDIA H200 NVL data center GPU, offering enterprises enhanced capabilities for agentic and generative AI use cases.
2. AI and Data Science Software Catalog
The AI and Data Science Software Catalog includes NVIDIA NIM microservices, frameworks such as NeMo, Riva, and Metropolis, as well as tools like Triton Inference Server, TensorRT, Tensor-LLM, PyTorch, and RAPIDS. These software packages are containerized for convenient deployment on various platforms.
3. Infrastructure Software Collection
The Infrastructure Software Collection comprises components necessary to support AI and data science software on accelerated systems. It includes GPU drivers, networking support, Kubernetes operators, and management essentials like Base Command Manager. The latest version of the collection introduces support for the H200 NVL GPU, offering improved performance and energy efficiency for AI workloads.
4. NVIDIA AI Enterprise Software Reference Architecture
A guide has been developed to assist organizations in quickly deploying and configuring on-premises servers to run a wide range of AI production applications. Systems sold with the H200 NVL GPU include a five-year NVIDIA AI Enterprise subscription, and various options are available for getting started with the platform.
Note: The summary has been provided in English, as requested.