docs.nvidia.com

# NVIDIA Technical Product Documentation

> Collection of NVIDIA Technical Documentation llms.txt files for AI user agents. Content is in English and listed in alphabetical order. Last updated 16 July 2026

## NVIDIA AI Cluster Runtime (AICR)

> Tooling for generating validated, optimized GPU-accelerated Kubernetes configurations through a snapshot-recipe-validate-bundle workflow.

- [NVIDIA AI Cluster Runtime](https://docs.nvidia.com/aicr/llms.txt)

## NVIDIA AIPerf

> AIPerf is a comprehensive CLI benchmarking tool that measures the performance of generative AI models across inference backends, delivering detailed real-time metrics and benchmark reports.

- [NVIDIA AIPerf](https://docs.nvidia.com/aiperf/llms.txt)

## NVIDIA AIStore

> AIStore is a deploy-anywhere distributed object store for AI workloads, with seamless fast-tiering for cloud storage and linear scalability.

- [NVIDIA AIStore](https://docs.nvidia.com/aistore/llms.txt)

## NVIDIA Brev

> NVIDIA Brev provides easy access to GPU sandboxes for experimentation.

- [NVIDIA Brev](https://docs.nvidia.com/brev/llms.txt)

## NVIDIA Clara Parabricks

> Parabricks is a free software suite for performing secondary analysis of next generation sequencing (NGS) DNA and RNA data.

- [NVIDIA Clara Parabricks](https://docs.nvidia.com/clara/parabricks/llms.txt)

## NVIDIA Cloud Functions (NVCF)

> A unified API layer for scaling inference and simulation workloads across one or more Kubernetes clusters.

- [NVIDIA Cloud Functions](https://docs.nvidia.com/nvcf/llms.txt)

## NVIDIA Cosmos

> Suite of models, ontologies, and benchmarks to enable multimodal LLMs to generate physically-grounded responses

- [NVIDIA Cosmos](https://docs.nvidia.com/cosmos/llms.txt)

## NVIDIA CUDA Toolkit

> Development environment for creating high performance GPU-accelerated applications with libraries, compilers, and debugging tools.

- [NVIDIA CUDA Toolkit](https://docs.nvidia.com/cuda/llms.txt)

## NVIDIA CUDA-Q

> A Platform for accelerated quantum supercomputing that offers a hybrid programming model for heterogeneous environments that include QPUs, GPUs and CPUs.

- [NVIDIA CUDA-Q documentation on GitHub](https://nvidia.github.io/cuda-quantum/latest/llms.txt)

## NVIDIA cuVS

> cuVS is NVIDIA's open-source GPU library for unstructured data processing: nearest neighbors, clustering, and vector compression

- [NVIDIA cuVS](https://docs.nvidia.com/cuvs/llms.txt)

## NVIDIA Dynamo

> High-throughput, low-latency inference framework designed for serving generative AI and reasoning models in multi-node distributed environments.

- [NVIDIA Dynamo](https://docs.nvidia.com/dynamo/llms.txt)

## NVIDIA HEAVY.AI

> Heavy.AI is a GPU-accelerated database and visualization platform that is optimized for interactive analytics of large-scale geospatial and time-series data.

- [NVIDIA HEAVY.AI](https://docs.nvidia.com/heavyai/llms.txt)

## NVIDIA Holoscan SDK User Guide

> Holoscan SDK is NVIDIA's platform for building and deploying real-time, AI-enabled sensor processing applications across edge, embedded, and cloud environments.

- [NVIDIA Holoscan SDK User Guide](https://docs.nvidia.com/holoscan/sdk-user-guide/llms.txt)

## NVIDIA Jetson Software

> Edge software platform for real-time AI and robotics supporting all NVIDIA Jetson hardware.

- [NVIDIA Jetson Software](https://docs.nvidia.com/jetson/llms.txt)

## NVIDIA NeMo Agent Toolkit

> Open-source library that provides framework-agnostic profiling, evaluation, and optimization for AI agent systems.

- [NVIDIA NeMo Agent Toolkit](https://docs.nvidia.com/nemo/agent-toolkit/llms.txt)

## NVIDIA NeMo AutoModel

> PyTorch DTensor-native SPMD training library for scaling training and fine-tuning of LLMs, VLMs, diffusion models, and retrieval models, with Hugging Face integration.

- [NVIDIA NeMo AutoModel](https://docs.nvidia.com/nemo/automodel/llms.txt)

## NVIDIA NeMo Curator

> Open-source, enterprise-grade toolkit for scalable, privacy-aware data curation across text, image, video, and audio modalities.

- [NVIDIA NeMo Curator](https://docs.nvidia.com/nemo/curator/llms.txt)

## NVIDIA NeMo Data Designer

> Orchestration framework for generating high-quality synthetic datasets from scratch or seed data, with batching, parallelism, validation, and reproducible workflows.

- [NVIDIA NeMo Data Designer](https://docs.nvidia.com/nemo/datadesigner/llms.txt)

## NVIDIA NeMo Framework

> Cloud-native framework to create, customize, and deploy new generative AI models by leveraging existing code and pretrained model checkpoints.

- [NVIDIA NeMo Framework](https://docs.nvidia.com/nemo-framework/llms.txt)

## NVIDIA NeMo Guardrails

> Open-source toolkit for easily adding programmable guardrails to LLM-based conversational systems.

- [NVIDIA NeMo Guardrails](https://docs.nvidia.com/nemo/guardrails/llms.txt)

## NVIDIA NeMo Gym

> Open-source library for evaluating and improving LLMs and agents with scalable RL environments, verified rollouts, and training-framework integrations.

- [NVIDIA NeMo Gym](https://docs.nvidia.com/nemo/gym/llms.txt)

## NVIDIA NeMo Microservices

> API-first modular set of tools to customize, evaluate, and secure large language and embedding models on-premise or in the cloud.

- [NVIDIA NeMo Microservices](https://docs.nvidia.com/nemo/microservices/llms.txt)

## NVIDIA NeMo Platform

> Platform that brings NVIDIA NeMo libraries together under one CLI, Python SDK, and web UI for hardening, evaluating, and tuning production AI agents.

- [NVIDIA NeMo Platform](https://docs.nvidia.com/nemo-platform/llms.txt)

## NVIDIA NeMo Relay

> Multi-language agent runtime that provides shared scopes, policy, plugins, lifecycle events, and middleware for tool and LLM calls across agent systems.

- [NVIDIA NeMo Relay](https://docs.nvidia.com/nemo/relay/llms.txt)

## NVIDIA NeMo Retriever

> Collection of microservices for building and scaling multimodal data extraction, embedding, and reranking pipelines.

- [NVIDIA NeMo Retriever](https://docs.nvidia.com/nemo/retriever/latest/llms.txt)

## NVIDIA NeMo RL

> Open-source post-training library to streamline and scale reinforcement learning methods for multimodal models (LLMs, VLMs etc).

- [NVIDIA NeMo RL](https://docs.nvidia.com/nemo/rl/llms.txt)

## NVIDIA NemoClaw

> Open source stack that simplifies running OpenClaw always-on assistants more safely, with a single command

- [NVIDIA NemoClaw](https://docs.nvidia.com/nemoclaw/llms.txt)

## NVIDIA Nemotron

> Family of open models with open weights, training data, and recipes for building specialized AI agents.

- [NVIDIA Nemotron Documentation on HuggingFace](https://huggingface.co/docs/transformers/en/model_doc/nemotron.md)

## NVIDIA NIM

> Containers to self-host GPU-accelerated inferencing microservices for pretrained and customized AI models.

- [NVIDIA NIM](https://docs.nvidia.com/nim/llms.txt)

## NVIDIA NVSentinel

> Kubernetes-native GPU cluster health monitoring and automated fault remediation framework.

- [NVIDIA NVSentinel](https://docs.nvidia.com/nvsentinel/llms.txt)

## NVIDIA OpenShell

> Open Source runtime for building and deploying autonomous, self-evolving agents more safely. safe, private runtime for autonomous AI agents with Sandboxes

- [NVIDIA OpenShell](https://docs.nvidia.com/openshell/llms.txt)

## NVIDIA Riva

> GPU-accelerated SDK for building Speech and Conversational AI applications.

- [NVIDIA Riva](https://docs.nvidia.com/deeplearning/riva/llms.txt)

## NVIDIA SDGM

> NVIDIA SDGM is a next-generation predictive AI platform for relational data that lets you generate high-quality predictions directly from your data warehouse without feature engineering.

- [NVIDIA SDGM](https://docs.nvidia.com/sdgm/llms.txt)

## NVIDIA Topograph

> Discovers physical network topology of GPU clusters and exposes it to workload schedulers (Slurm, Kubernetes, Slinky).

- [NVIDIA Topograph](https://docs.nvidia.com/topograph/llms.txt)