The NVIDIA NeMo Agent toolkit is an open-source library for efficiently connecting and optimizing teams of AI agents.

The developer-first platform for scaling complex Physical AI workloads across heterogeneous compute—unifying training GPUs, simulation clusters, and edge devices in a simple YAML

`std::execution`, the proposed C++ framework for asynchronous and parallel programming.

NVIDIA Bare Metal Manager - Hardware Lifecycle Management and multitenant networking

A unified library of SOTA model optimization techniques like quantization, pruning, distillation, speculative decoding, etc. It compresses deep learning models for downstream deployment frameworks like TensorRT-LLM, TensorRT, vLLM, etc. to optimize inference speed.

Ongoing research training transformer models at scale

NVIDIA Bare Metal Management - Hardware Lifeceycle managment (REST API)

C++ and Python support for the CUDA Quantum programming model for heterogeneous quantum-classical workflows

BioNeMo Framework: For building and adapting AI models in drug discovery at scale

NVIDIA GPU Operator creates, configures, and manages GPUs in Kubernetes