Application Layer Software

Application Layer Components: The application layer provides specialized SDKs, frameworks, optimized microservices, and pre-trained models. It includes development and deployment tools such as NVIDIA Triton Inference Server, TensorFlow, and PyTorch, along with optimized libraries for deep learning, data science, and machine learning.

AI Solution Development: These components enable building and deploying AI solutions — generative AI, AI agents, physical AI, and domain-specific applications. By separating the infrastructure layer (which is versioned) from the application layer, NVIDIA AI Enterprise ensures that foundational updates and improvements do not disrupt the development and deployment of AI applications. For infrastructure layer components, refer to Infrastructure Layer Software.

Release Distribution: Application software is distributed through three release branch types: Feature Branch (FB) for latest innovation, Production Branch (PB) for production stability, and Long-Term Support Branch (LTSB) for regulated environments requiring 36 months of API stability. For more information, refer to the Lifecycle Policy.

Software Components

NVIDIA AI Enterprise includes the following NVIDIA application layer software components. For infrastructure layer components, refer to Infrastructure Layer Software.

Attention

This is a representative list of key software components and is not exhaustive. The full set of supported software evolves with each release. Refer to the NVIDIA NGC Catalog: Application Layer Software Components the individual product documentation, and the release notes for the most current information.

Table 2 Application Layer Software Components

Component	Description	NGC Catalog	Documentation
NGC Catalog Resources
NVIDIA AI Enterprise	Lists all NVIDIA AI Enterprise supported software available on NGC, including AI frameworks, microservices, pre-trained models, SDKs, and tools with enterprise support.	NVIDIA AI Enterprise on NGC	NVIDIA AI Enterprise Docs Hub
Inference & Deployment
NVIDIA NIM	Optimized microservices for accelerating foundation model deployment on any cloud or data center. Includes production-grade runtimes, security updates, and API references.	NVIDIA NIM on NGC	NVIDIA NIM Documentation
NVIDIA TensorRT	High-performance deep learning inference optimizer and runtime for production deployment of trained models.	NVIDIA TensorRT on NGC	NVIDIA TensorRT Documentation
NVIDIA Triton Inference Server	Multi-framework inference server with optimized backends for deploying AI models at scale. Supports multiple model formats and dynamic batching.	NVIDIA Triton on NGC	NVIDIA Triton Documentation
AI Frameworks & Libraries
NVIDIA CUDA Deep Learning	Container bundle with GPU-accelerated deep learning libraries including cuDNN, NCCL, and CUDA runtime for training and inference workloads.	NVIDIA CUDA DL on NGC	NVIDIA CUDA DL Release Notes
NVIDIA NeMo	End-to-end platform for building, customizing, and deploying generative AI models including LLMs, multimodal, speech AI, and vision.	NVIDIA NeMo on NGC	NVIDIA NeMo Documentation
PyTorch	NVIDIA-optimized PyTorch container with GPU-accelerated deep learning and data science libraries.	PyTorch on NGC	PyTorch Release Notes
NVIDIA RAPIDS	GPU-accelerated data science libraries for data preparation, machine learning, and graph analytics.	NVIDIA RAPIDS on NGC	NVIDIA RAPIDS Documentation
NVIDIA RAPIDS Accelerator for Apache Spark	GPU acceleration plugin for Apache Spark 3 data science pipelines and AI model training.	NVIDIA RAPIDS Spark on NGC	NVIDIA RAPIDS Accelerator Documentation
Domain SDKs & Toolkits
NVIDIA Clara Parabricks	GPU-accelerated computational genomics toolkit for secondary and tertiary analysis of next-generation sequencing data.	NVIDIA Clara Parabricks on NGC	NVIDIA Clara Parabricks Documentation
NVIDIA DeepStream SDK	Streaming analytics toolkit for building AI-powered video and sensor data applications at scale.	NVIDIA DeepStream on NGC	NVIDIA DeepStream Documentation
NVIDIA Holoscan SDK	Sensor processing platform for building real-time AI pipelines in healthcare, industrial inspection, and edge applications.	NVIDIA Holoscan on NGC	NVIDIA Holoscan Documentation
NVIDIA Morpheus · End of Support — refer to EOL Notices	Cybersecurity AI framework for real-time threat detection and digital fingerprinting at data center scale.	NVIDIA Morpheus on NGC	NVIDIA Morpheus Documentation
NVIDIA Omniverse	Provides a platform for 3D collaboration, simulation, and content creation with real-time collaboration and photorealistic rendering capabilities.	NVIDIA Omniverse on NGC	NVIDIA Omniverse Documentation
NVIDIA Riva	GPU-accelerated speech AI and conversational AI services for automatic speech recognition (ASR), text-to-speech (TTS), and natural language processing.	NVIDIA Riva on NGC	NVIDIA Riva Documentation
NVIDIA TAO Toolkit	AI model adaptation toolkit for fine-tuning pre-trained models with custom data, enabling transfer learning workflows.	NVIDIA TAO on NGC	NVIDIA TAO Documentation
Pre-trained Models
Production-Ready Pre-trained Models	Pre-trained AI models that simplify and accelerate development by eliminating the need to build from scratch.	Pre-trained Models on NGC	Varies by model. Refer to the documentation links on the product pages on NGC Catalog.

Release Branches and Lifecycle

Application software is distributed through three release branch types — Feature Branch (FB), Production Branch (PB), and Long-Term Support Branch (LTSB) — each offering a different balance of innovation speed, API stability, and support duration.

• NVIDIA AI Enterprise Lifecycle Policy — Defines each branch type, support periods, and update cadence.

• Choosing the Right Release Branch — Decision guide with comparison table and industry scenarios.

• Application Software Releases — Active and archived release branches.