#gpu

cuDF - GPU DataFrame Library

Ongoing research training transformer models at scale

CUDA Core Compute Libraries

NVIDIA cuML: GPU-Accelerated Machine Learning

A Python framework for GPU-accelerated simulation, robotics, and machine learning.

Open ABI and FFI for Machine Learning Systems

NumPy & SciPy for GPU

A cross-platform, safe, pure-Rust graphics API.

On-device AI across mobile, embedded and edge for PyTorch

Tensors and Dynamic neural networks in Python with strong GPU acceleration

NVIDIA RMM is a library for allocating and managing GPU memory in C++ and Python.

FlashInfer: Kernel Library for LLM Serving

The Triton Inference Server provides an optimized cloud and edge inferencing solution.

🌊 Julia software for fast, friendly, flexible, ocean-flavored fluid dynamics on CPUs and GPUs

NVSentinel detects and remediates GPU faults on Kubernetes nodes

Tooling for optimized, validated, and reproducible GPU-accelerated AI runtime in Kubernetes

cuDNN Frontend is NVIDIA's modern, open-source entry point to the cuDNN library and a growing collection of high-performance open-source kernels.

Simulate NVIDIA infrastructure (e.g. GPU) on CPU nodes in Kubernetes :coffee:️

RAFT contains fundamental widely-used algorithms and primitives for machine learning and information retrieval. The algorithms are CUDA-accelerated and form building blocks for more easily writing high performance applications.

Accessible large language models via k-bit quantization for PyTorch.

DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective.

Platform for deploying and routing GPU-accelerated inference, streaming, and batch workloads at scale.

Offline optimization of your disaggregated Dynamo graph

PyGraphistry is a Python library to quickly load, shape, embed, and explore big graphs with the GPU-accelerated Graphistry visual graph analyzer

Time series forecasting with PyTorch

The AI Compute Platform for frontier teams. SkyPilot turns fragmented AI compute into one AI supercomputer, so frontier AI teams build custom intelligence faster.

Burn is a next generation tensor library and Deep Learning Framework that doesn't compromise on flexibility, efficiency and portability.

cuTile is a programming model for writing parallel kernels for NVIDIA GPUs

CUDA Templates and Python DSLs for High-Performance Linear Algebra

Main repository for QMCPACK, an open-source production level many-body ab initio Quantum Monte Carlo code for computing the electronic structure of atoms, molecules, and solids with full performance portable GPU support