This package provides the proprietary cuTENSOR library for NVIDIA GPUs.

This package provides the CUDA run-time support libraries for NVIDIA GPUs, all of which are proprietary.

This package provides the NVIDIA tool for debugging CUDA applications running. CUDA-GDB is an extension to GDB, the GNU Project debugger. The tool provides developers with a mechanism for debugging CUDA applications running on actual hardware. This enables developers to debug applications without the potential variations introduced by simulation and emulation environments.

This package provides the CUDA Direct Sparse Solver library.

This package provides the CUDA Deep Neural Network library.

This package provides a set of GPU-accelerated basic linear algebra subroutines used for handling sparse matrices that perform significantly faster than CPU-only alternatives. Depending on the specific operation, the library targets matrices with sparsity ratios in the range between 70%-99.9%.

This package provides the CUDA compiler and the CUDA run-time support libraries for NVIDIA GPUs, all of which are proprietary.

This package provides a cross-platform API for annotating source code to provide contextual information to developer tools.

This package provides the CUDA Deep Neural Network library.

CUTLASS is a collection of CUDA C++ template abstractions for implementing high-performance matrix-matrix multiplication (GEMM) and related computations at all levels and scales within CUDA. It incorporates strategies for hierarchical decomposition and data movement similar to those used to implement cuBLAS and cuDNN.

CUTLASS decomposes these ``moving parts'' into reusable, modular software components abstracted by C++ template classes. Primitives for different levels of a conceptual parallelization hierarchy can be specialized and tuned via custom tiling sizes, data types, and other algorithmic policy. The resulting flexibility simplifies their use as building blocks within custom kernels and applications.

NCCL (pronounced "Nickel") is a stand-alone library of standard communication routines for NVIDIA GPUs, implementing all-reduce, all-gather, reduce, broadcast, reduce-scatter, as well as any send/receive based communication pattern. It has been optimized to achieve high bandwidth on platforms using PCIe, NVLink, NVswitch, as well as networking using InfiniBand Verbs or TCP/IP sockets. NCCL supports an arbitrary number of GPUs installed in a single node or across multiple nodes, and can be used in either single- or multi-process (e.g., MPI) applications.

This package provides a an interactive profiler for CUDA and NVIDIA OptiX that provides detailed performance metrics and API debugging via a user interface and command-line tool. Users can run guided analysis and compare results with a customizable and data-driven user interface, as well as post-process and analyze results in their own workflows.

This package provides a GPU-accelerated library of primitives for deep neural networks, with highly tuned implementations for standard routines such as forward and backward convolution, attention, matmul, pooling, and normalization.

This package provides the CUDA compiler and the CUDA run-time support libraries for NVIDIA GPUs, all of which are proprietary.

This package provides a functional correctness checking suite included in the CUDA toolkit. This suite contains multiple tools that can perform different type of checks. The memcheck tool is capable of precisely detecting and attributing out of bounds and misaligned memory access errors in CUDA applications, and can also report hardware exceptions encountered by the GPU. The racecheck tool can report shared memory data access hazards that can cause data races. The initcheck tool can report cases where the GPU performs uninitialized accesses to global memory. The synccheck tool can report cases where the application is attempting invalid usages of synchronization primitives.

This package provides a system-wide performance analysis tool designed to visualize an application’s algorithms, identify the largest opportunities to optimize, and tune to scale efficiently across any quantity or size of CPUs and GPUs,from large servers to small systems-on-a-chip.

This package provides a GPU-accelerated library of primitives for deep neural networks, with highly tuned implementations for standard routines such as forward and backward convolution, attention, matmul, pooling, and normalization.

This package provides a set of APIs which can be used at runtime to combine multiple CUDA objects into one CUDA fat binary (fatbin). The APIs accept inputs in multiple formats, either device cubins, PTX, or LTO-IR. The output is a fatbin that can be loaded by cuModuleLoadData of the CUDA Driver API. The functionality in this library is similar to the fatbinary offline tool in the CUDA toolkit, with the following advantages:

• Support for runtime fatbin creation.

• The clients get fine grain control over the input process.

• Supports direct input from memory, rather than requiring inputs be written to files.

This package provides the CUDA compiler and the CUDA run-time support libraries for NVIDIA GPUs, all of which are proprietary.

This package provides facilities that focus on the simple and efficient generation of high-quality pseudorandom and quasirandom numbers. A pseudorandom sequence of numbers satisfies most of the statistical properties of a truly random sequence but is generated by a deterministic algorithm. A quasirandom sequence of -dimensional points is generated by a deterministic algorithm designed to fill an -dimensional space evenly.

This package provides the CUDA compiler and the CUDA run-time support libraries for NVIDIA GPUs, all of which are proprietary.

This package enables the creation of profiling and tracing tools that target CUDA applications and give insight into the CPU and GPU behavior of CUDA applications. It provides the following APIs:

• the Activity API,

• the Callback API,

• the Event API,

• the Metric API,

• the Profiling API,

• the PC Sampling API,

• the Checkpoint API.

This package provides a high-level library based on the cuBLAS and cuSPARSE libraries. It consists of two modules corresponding to two sets of API: the cuSolver API on a single GPU; and the cuSolverMG API on a single node multiGPU. Each of these can be used independently or in concert with other toolkit libraries. The intent of cuSolver is to provide useful LAPACK-like features, such as common matrix factorization and triangular solve routines for dense matrices, a sparse least-squares solver and an eigenvalue solver. In addition, cuSolver provides a new refactorization library useful for solving sequences of matrices with a shared sparsity pattern.

CUTLASS is a collection of CUDA C++ template abstractions for implementing high-performance matrix-matrix multiplication (GEMM) and related computations at all levels and scales within CUDA. It incorporates strategies for hierarchical decomposition and data movement similar to those used to implement cuBLAS and cuDNN.

CUTLASS decomposes these ``moving parts'' into reusable, modular software components abstracted by C++ template classes. Primitives for different levels of a conceptual parallelization hierarchy can be specialized and tuned via custom tiling sizes, data types, and other algorithmic policy. The resulting flexibility simplifies their use as building blocks within custom kernels and applications.