https://conda.anaconda.org/nvidia The most recent 100 updates for nvidia. Wed, 24 Jun 2026 20:55:21 GMT Wed, 24 Jun 2026 20:55:21 GMT The NVIDIA Collective Communications Library (NCCL) implements multi-GPU and multi-node collective communication primitives that are performance optimized for NVIDIA GPUs. NCCL provides routines such as all-gather, all-reduce, broadcast, reduce, reduce-scatter, that are optimized to achieve high bandwidth over PCIe and NVLink high-speed interconnect. https://docs.nvidia.com/deeplearning/sdk/nccl-developer-guide/docs/index.html https://github.com/NVIDIA/nccl Wed, 10 Jun 2026 17:20:14 GMT https://developer.nvidia.com/nccl A Python client and command-line utility for CuOpt managed service Tue, 09 Jun 2026 22:46:44 GMT https://docs.nvidia.com/cuopt/introduction.html cuOpt Microserver - GPU Combinatorial Optimization Tue, 09 Jun 2026 22:46:39 GMT https://docs.nvidia.com/cuopt/introduction.html cuOpt - GPU Combinatorial Optimization Tue, 09 Jun 2026 22:44:21 GMT https://docs.nvidia.com/cuopt/introduction.html cuOpt - GPU Combinatorial Optimization Tue, 09 Jun 2026 22:09:25 GMT https://docs.nvidia.com/cuopt/introduction.html libcuopt test & benchmark executables Tue, 09 Jun 2026 22:09:25 GMT https://docs.nvidia.com/cuopt/introduction.html NVIDIA CUDA Deep Neural Network (cuDNN) is a GPU-accelerated library of primitives for deep neural networks. It provides highly tuned implementations of routines arising frequently in DNN applications. License Agreements:- The packages are governed by the NVIDIA cuDNN Software License Agreement (EULA). By downloading and using the packages, you accept the terms and conditions of the NVIDIA cuDNN EULA - https://docs.nvidia.com/deeplearning/cudnn/sla/index.html https://docs.nvidia.com/deeplearning/cudnn/ https://developer.nvidia.com/rdp/cudnn-download Mon, 08 Jun 2026 13:20:59 GMT https://developer.nvidia.com/cudnn NVIDIA CUDA Deep Neural Network (cuDNN) is a GPU-accelerated library of primitives for deep neural networks. It provides highly tuned implementations of routines arising frequently in DNN applications. License Agreements:- The packages are governed by the NVIDIA cuDNN Software License Agreement (EULA). By downloading and using the packages, you accept the terms and conditions of the NVIDIA cuDNN EULA - https://docs.nvidia.com/deeplearning/cudnn/sla/index.html https://docs.nvidia.com/deeplearning/cudnn/ https://developer.nvidia.com/rdp/cudnn-download Mon, 08 Jun 2026 13:17:12 GMT https://developer.nvidia.com/cudnn No description. Sun, 07 Jun 2026 01:57:06 GMT cuTENSOR is a high-performance CUDA library for tensor primitives. https://docs.nvidia.com/cuda/cutensor/index.html Fri, 05 Jun 2026 13:55:55 GMT https://developer.nvidia.com/cutensor cuTENSOR is a high-performance CUDA library for tensor primitives. https://docs.nvidia.com/cuda/cutensor/index.html Fri, 05 Jun 2026 13:51:58 GMT https://developer.nvidia.com/cutensor The NVIDIA Performance Libraries (NVPL). For convenience only! Not to be used in host requirements for conda recipes! Wed, 27 May 2026 22:13:37 GMT https://developer.nvidia.com/nvpl The NVIDIA Performance Libraries (NVPL). For convenience only! Not to be used in host requirements for conda recipes! Wed, 27 May 2026 22:13:36 GMT https://developer.nvidia.com/nvpl The NVIDIA Performance Libraries (NVPL) are a collection of high performance mathematical libraries optimized for the NVIDIA Grace Armv9.0 architecture. These CPU-only libraries have no dependencies on CUDA or CTK, and are drop in replacements for standard C and Fortran mathematical APIs allowing HPC applications to achieve maximum performance on the Grace platform. https://docs.nvidia.com/nvpl/ Wed, 27 May 2026 21:42:39 GMT https://developer.nvidia.com/nvpl The NVIDIA Performance Libraries (NVPL) are a collection of high performance mathematical libraries optimized for the NVIDIA Grace Armv9.0 architecture. These CPU-only libraries have no dependencies on CUDA or CTK, and are drop in replacements for standard C and Fortran mathematical APIs allowing HPC applications to achieve maximum performance on the Grace platform. https://docs.nvidia.com/nvpl/ Wed, 27 May 2026 21:41:46 GMT https://developer.nvidia.com/nvpl The NVIDIA Performance Libraries (NVPL) are a collection of high performance mathematical libraries optimized for the NVIDIA Grace Armv9.0 architecture. These CPU-only libraries have no dependencies on CUDA or CTK, and are drop in replacements for standard C and Fortran mathematical APIs allowing HPC applications to achieve maximum performance on the Grace platform. https://docs.nvidia.com/nvpl/ Wed, 27 May 2026 21:30:20 GMT https://developer.nvidia.com/nvpl The NVIDIA Performance Libraries (NVPL) are a collection of high performance mathematical libraries optimized for the NVIDIA Grace Armv9.0 architecture. These CPU-only libraries have no dependencies on CUDA or CTK, and are drop in replacements for standard C and Fortran mathematical APIs allowing HPC applications to achieve maximum performance on the Grace platform. https://docs.nvidia.com/nvpl/ Wed, 27 May 2026 21:28:41 GMT https://developer.nvidia.com/nvpl The NVIDIA Performance Libraries (NVPL) are a collection of high performance mathematical libraries optimized for the NVIDIA Grace Armv9.0 architecture. These CPU-only libraries have no dependencies on CUDA or CTK, and are drop in replacements for standard C and Fortran mathematical APIs allowing HPC applications to achieve maximum performance on the Grace platform. https://docs.nvidia.com/nvpl/ Wed, 27 May 2026 21:15:51 GMT https://developer.nvidia.com/nvpl The NVIDIA Performance Libraries (NVPL) are a collection of high performance mathematical libraries optimized for the NVIDIA Grace Armv9.0 architecture. These CPU-only libraries have no dependencies on CUDA or CTK, and are drop in replacements for standard C and Fortran mathematical APIs allowing HPC applications to achieve maximum performance on the Grace platform. https://docs.nvidia.com/nvpl/ Wed, 27 May 2026 21:12:35 GMT https://developer.nvidia.com/nvpl The NVIDIA Performance Libraries (NVPL) are a collection of high performance mathematical libraries optimized for the NVIDIA Grace Armv9.0 architecture. These CPU-only libraries have no dependencies on CUDA or CTK, and are drop in replacements for standard C and Fortran mathematical APIs allowing HPC applications to achieve maximum performance on the Grace platform. https://docs.nvidia.com/nvpl/ Wed, 27 May 2026 21:00:29 GMT https://developer.nvidia.com/nvpl The NVIDIA Performance Libraries (NVPL) are a collection of high performance mathematical libraries optimized for the NVIDIA Grace Armv9.0 architecture. These CPU-only libraries have no dependencies on CUDA or CTK, and are drop in replacements for standard C and Fortran mathematical APIs allowing HPC applications to achieve maximum performance on the Grace platform. https://docs.nvidia.com/nvpl/ Wed, 27 May 2026 20:59:57 GMT https://developer.nvidia.com/nvpl The NVIDIA Performance Libraries (NVPL) are a collection of high performance mathematical libraries optimized for the NVIDIA Grace Armv9.0 architecture. These CPU-only libraries have no dependencies on CUDA or CTK, and are drop in replacements for standard C and Fortran mathematical APIs allowing HPC applications to achieve maximum performance on the Grace platform. https://docs.nvidia.com/nvpl/ Wed, 27 May 2026 20:42:50 GMT https://developer.nvidia.com/nvpl The NVIDIA Performance Libraries (NVPL) are a collection of high performance mathematical libraries optimized for the NVIDIA Grace Armv9.0 architecture. These CPU-only libraries have no dependencies on CUDA or CTK, and are drop in replacements for standard C and Fortran mathematical APIs allowing HPC applications to achieve maximum performance on the Grace platform. https://docs.nvidia.com/nvpl/ Wed, 27 May 2026 20:41:18 GMT https://developer.nvidia.com/nvpl The NVIDIA Performance Libraries (NVPL) are a collection of high performance mathematical libraries optimized for the NVIDIA Grace Armv9.0 architecture. These CPU-only libraries have no dependencies on CUDA or CTK, and are drop in replacements for standard C and Fortran mathematical APIs allowing HPC applications to achieve maximum performance on the Grace platform. https://docs.nvidia.com/nvpl/ Wed, 27 May 2026 20:31:37 GMT https://developer.nvidia.com/nvpl The NVIDIA Performance Libraries (NVPL) are a collection of high performance mathematical libraries optimized for the NVIDIA Grace Armv9.0 architecture. These CPU-only libraries have no dependencies on CUDA or CTK, and are drop in replacements for standard C and Fortran mathematical APIs allowing HPC applications to achieve maximum performance on the Grace platform. https://docs.nvidia.com/nvpl/ Wed, 27 May 2026 20:30:48 GMT https://developer.nvidia.com/nvpl The NVIDIA Performance Libraries (NVPL) are a collection of high performance mathematical libraries optimized for the NVIDIA Grace Armv9.0 architecture. These CPU-only libraries have no dependencies on CUDA or CTK, and are drop in replacements for standard C and Fortran mathematical APIs allowing HPC applications to achieve maximum performance on the Grace platform. https://docs.nvidia.com/nvpl/ Wed, 27 May 2026 20:16:27 GMT https://developer.nvidia.com/nvpl The NVIDIA Performance Libraries (NVPL) are a collection of high performance mathematical libraries optimized for the NVIDIA Grace Armv9.0 architecture. These CPU-only libraries have no dependencies on CUDA or CTK, and are drop in replacements for standard C and Fortran mathematical APIs allowing HPC applications to achieve maximum performance on the Grace platform. https://docs.nvidia.com/nvpl/ Wed, 27 May 2026 20:13:31 GMT https://developer.nvidia.com/nvpl The NVIDIA Performance Libraries (NVPL) are a collection of high performance mathematical libraries optimized for the NVIDIA Grace Armv9.0 architecture. These CPU-only libraries have no dependencies on CUDA or CTK, and are drop in replacements for standard C and Fortran mathematical APIs allowing HPC applications to achieve maximum performance on the Grace platform. https://docs.nvidia.com/nvpl/ Wed, 27 May 2026 19:46:47 GMT https://developer.nvidia.com/nvpl NVIDIA cublasMp is a high performance, multi-process, GPU accelerated library for distributed basic dense linear algebra. cuBLASMp is compatible with 2D block-cyclic data layout and provides PBLAS-like C APIs. https://docs.nvidia.com/cuda/index.html Wed, 20 May 2026 18:20:01 GMT https://developer.nvidia.com/cuda-toolkit Meta-package containing all the available packages required for libcufile. https://docs.nvidia.com/cuda/index.html Tue, 19 May 2026 19:52:30 GMT https://developer.nvidia.com/cuda-toolkit A meta-package for pinning to a CUDA release version https://docs.nvidia.com/cuda/index.html Tue, 19 May 2026 19:44:15 GMT https://developer.nvidia.com/cuda-toolkit NVIDIA cuBLASMp is a high performance, multi-process, GPU accelerated library for distributed basic dense linear algebra. cuBLASMp is compatible with 2D block-cyclic data layout and provides PBLAS-like C APIs. https://docs.nvidia.com/cuda/cublasmp/ Tue, 19 May 2026 15:20:50 GMT https://docs.nvidia.com/cuda/cublasmp/ NVIDIA cuBLASMp is a high performance, multi-process, GPU accelerated library for distributed basic dense linear algebra. cuBLASMp is compatible with 2D block-cyclic data layout and provides PBLAS-like C APIs. https://docs.nvidia.com/cuda/cublasmp/ Tue, 19 May 2026 15:20:01 GMT https://docs.nvidia.com/cuda/cublasmp/ Fast, GPU-powered components for electronic structure theory https://docs.nvidia.com/cuda/cuest/ Thu, 07 May 2026 18:36:32 GMT https://developer.nvidia.com/cuest Fast, GPU-powered components for electronic structure theory https://docs.nvidia.com/cuda/cuest/ Thu, 07 May 2026 18:35:40 GMT https://developer.nvidia.com/cuest Fast, GPU-powered components for electronic structure theory https://docs.nvidia.com/cuda/cuest/ Thu, 07 May 2026 18:33:44 GMT https://developer.nvidia.com/cuest This is a runtime package only. Developers should install libcudss-dev to build with cuDSS. Tue, 05 May 2026 21:48:15 GMT This is a runtime package only. Developers should install libcudss-dev to build with cuDSS. Tue, 05 May 2026 21:46:45 GMT NVIDIA cuDSS is an optimized, first-generation GPU-accelerated Direct Sparse Solver library for solving linear systems with sparse matrices. Direct Sparse Solvers are an important part of numerical computing as they provide a general robust way of solving large linear systems without and are capable of taking advantage of both high compute throughput and memory bandwidth of the GPUs. https://docs.nvidia.com/cuda/cudss/ Tue, 05 May 2026 21:41:33 GMT https://developer.nvidia.com/cudss This is a runtime package only. Developers should install libcudss-dev to build with cuDSS. Tue, 05 May 2026 21:40:18 GMT With Tile IR, we introduce a new operation set and programming model to retain CUDA’s performance across architectures while regaining portability and improving productivity for developers using matrix operations on new architectures. We virtualize tensor-cores and their associated programming model to the point that we can innovate new approaches in hardware without invalidating investments in software. https://docs.nvidia.com/cuda/index.html Sat, 02 May 2026 04:22:00 GMT https://developer.nvidia.com/cuda-toolkit Compiler for CUDA applications. https://docs.nvidia.com/cuda/index.html Sat, 25 Apr 2026 06:27:26 GMT https://developer.nvidia.com/cuda-toolkit Compiler for CUDA applications. https://docs.nvidia.com/cuda/index.html Sat, 25 Apr 2026 06:25:41 GMT https://developer.nvidia.com/cuda-toolkit Compiler for CUDA applications. https://docs.nvidia.com/cuda/index.html Sat, 25 Apr 2026 04:37:46 GMT https://developer.nvidia.com/cuda-toolkit Compiler for CUDA applications. https://docs.nvidia.com/cuda/index.html Sat, 25 Apr 2026 04:18:56 GMT https://developer.nvidia.com/cuda-toolkit The CUDA cuLIBOS static library is required to link against the static CUDA math libraries. https://docs.nvidia.com/cuda/index.html Sat, 25 Apr 2026 03:23:36 GMT https://developer.nvidia.com/cuda-toolkit cuDLA is an extension of the CUDA programming model to encompass DLA work as well. It provides familiar CUDA programming model to work with DLA, reduces SW complexity (without need for EGLStreams) to interoperate between DLA – GPU for your inference pipelines. https://docs.nvidia.com/cuda/index.html Sat, 18 Apr 2026 04:04:53 GMT https://developer.nvidia.com/cuda-toolkit cuDLA is an extension of the CUDA programming model to encompass DLA work as well. It provides familiar CUDA programming model to work with DLA, reduces SW complexity (without need for EGLStreams) to interoperate between DLA – GPU for your inference pipelines. https://docs.nvidia.com/cuda/index.html Sat, 18 Apr 2026 04:04:07 GMT https://developer.nvidia.com/cuda-toolkit NVIDIA cuSPARSELt is a high-performance CUDA library dedicated to general matrix-matrix operations in which at least one operand is a structured sparse matrix with 50\% sparsity. The cuSPARSELt APIs allow flexibility in the algorithm/operation selection, epilogue, and matrix characteristics, including memory layout, alignment, and data types. License Agreements:- The packages are governed by the NVIDIA cuSPARSELt Software License Agreement (EULA). By downloading and using the packages, you accept the terms and conditions of the NVIDIA cuSPARSELt EULA - https://docs.nvidia.com/cuda/cusparselt/license.html https://docs.nvidia.com/cuda/cusparselt/index.html https://developer.nvidia.com/cusparselt/downloads Fri, 17 Apr 2026 07:04:58 GMT https://developer.nvidia.com/cusparse This is a runtime package only. Developers should install cusparselt-dev to build with cuSPARSELt. Fri, 17 Apr 2026 07:03:45 GMT cuOpt - GPU Combinatorial Optimization Thu, 09 Apr 2026 21:21:47 GMT https://docs.nvidia.com/cuopt/introduction.html cuOpt - GPU Combinatorial Optimization Thu, 09 Apr 2026 20:43:46 GMT https://docs.nvidia.com/cuopt/introduction.html cuObject (GPUDirect Storage for Objects) is a high performance suite of libraries designed to enable direct data transfers between GPU memory or system memory and S3 compatible object storage using RDMA https://docs.nvidia.com/gpudirect-storage/cuobject/index.html Thu, 09 Apr 2026 05:57:20 GMT https://developer.nvidia.com/cuda-toolkit cuObject (GPUDirect Storage for Objects) is a high performance suite of libraries designed to enable direct data transfers between GPU memory or system memory and S3 compatible object storage using RDMA https://docs.nvidia.com/gpudirect-storage/cuobject/index.html Thu, 09 Apr 2026 05:56:32 GMT https://developer.nvidia.com/cuda-toolkit No description. Mon, 06 Apr 2026 21:32:24 GMT NVIDIA cuSOLVERMp is a high-performance, distributed-memory, GPU-accelerated library that provides tools for the solution of dense linear systems and eigenvalue problems. cuSOLVERMp is compatible with 2D block-cyclic data layout and provides ScaLAPACK-like C APIs. A companion library, CAL, contains utilities to manage communicators and to synchronize processes in a safe way. https://docs.nvidia.com/cuda/index.html Mon, 06 Apr 2026 21:22:02 GMT https://developer.nvidia.com/cuda-toolkit NVIDIA cuQuantum SDK is a high-performance library for quantum information science and beyond. Currently its primary target is quantum circuit simulations and it contains the major component: - cuStateVec: a high-performance library for state vector computations - cuTensorNet: a high-performance library for tensor network computations License Agreements:- The packages are governed by the NVIDIA cuQuantum Software License Agreement (EULA). By downloading and using the packages, you accept the terms and conditions of the NVIDIA cuQuantum EULA - https://docs.nvidia.com/cuda/cuquantum/license.html https://docs.nvidia.com/cuda/cuquantum/index.html https://developer.nvidia.com/cuquantum-downloads Mon, 06 Apr 2026 21:02:17 GMT https://developer.nvidia.com/cuquantum-sdk NVIDIA cuSOLVERMp is a high-performance, distributed-memory, GPU-accelerated library that provides tools for the solution of dense linear systems and eigenvalue problems. cuSOLVERMp is compatible with 2D block-cyclic data layout and provides ScaLAPACK-like C APIs. https://docs.nvidia.com/cuda/cusolvermp/ Mon, 06 Apr 2026 06:17:01 GMT https://docs.nvidia.com/cuda/cusolvermp/ NVIDIA cuSOLVERMp is a high-performance, distributed-memory, GPU-accelerated library that provides tools for the solution of dense linear systems and eigenvalue problems. cuSOLVERMp is compatible with 2D block-cyclic data layout and provides ScaLAPACK-like C APIs. https://docs.nvidia.com/cuda/cusolvermp/ Mon, 06 Apr 2026 06:14:12 GMT https://docs.nvidia.com/cuda/cusolvermp/ NVIDIA NVSHMEM is an NVIDIA based "shared memory" library that provides an easy-to-use CPU-side interface to allocate pinned memory that is symmetrically distributed across a cluster of NVIDIA GPUs. NVSHMEM can significantly reduce communication and coordination overheads by allowing programmers to perform these operations from within CUDA kernels and on CUDA streams. https://docs.nvidia.com/nvshmem/api/index.html Sat, 21 Mar 2026 21:59:26 GMT https://docs.nvidia.com/nvshmem/index.html NVIDIA NVSHMEM is an NVIDIA based "shared memory" library that provides an easy-to-use CPU-side interface to allocate pinned memory that is symmetrically distributed across a cluster of NVIDIA GPUs. NVSHMEM can significantly reduce communication and coordination overheads by allowing programmers to perform these operations from within CUDA kernels and on CUDA streams. https://docs.nvidia.com/nvshmem/api/index.html Sat, 21 Mar 2026 21:41:37 GMT https://docs.nvidia.com/nvshmem/index.html NVIDIA NVSHMEM is an NVIDIA based "shared memory" library that provides an easy-to-use CPU-side interface to allocate pinned memory that is symmetrically distributed across a cluster of NVIDIA GPUs. NVSHMEM can significantly reduce communication and coordination overheads by allowing programmers to perform these operations from within CUDA kernels and on CUDA streams. https://docs.nvidia.com/nvshmem/api/index.html Sat, 21 Mar 2026 21:31:19 GMT https://docs.nvidia.com/nvshmem/index.html C++ accelerated Python bindings for NVIDIA NVSHMEM built with Cython. NVIDIA NVSHMEM is an NVIDIA based "shared memory" library that provides an easy-to-use CPU-side interface to allocate pinned memory that is symmetrically distributed across a cluster of NVIDIA GPUs. https://docs.nvidia.com/nvshmem/api/index.html Sat, 21 Mar 2026 20:09:28 GMT https://docs.nvidia.com/nvshmem/index.html CUDA Compile Time Advisor (ctadvisor) analyzes trace files containing compilation time information generated by NVCC or NVRTC. ctadvisor identifies compilation bottlenecks that take significant amount of time, for examples, template instantiation and headers processing. The tool provides users with suggestions to reduce compilation time. https://docs.nvidia.com/cuda/index.html Fri, 20 Mar 2026 07:12:50 GMT https://developer.nvidia.com/cuda-toolkit Python module supporting document ingestion. Wed, 21 Jan 2026 14:13:21 GMT https://github.com/NVIDIA/nv-ingest Python module supporting document ingestion. Wed, 21 Jan 2026 14:11:49 GMT https://github.com/NVIDIA/nv-ingest Python module with core document ingestion functions. Wed, 21 Jan 2026 14:09:38 GMT https://github.com/NVIDIA/nv-ingest CUDA Python provides a standard set of low-level interfaces, providing full coverage of and access to the CUDA host APIs from Python. https://nvidia.github.io/cuda-python/cuda-bindings https://github.com/NVIDIA/cuda-python https://github.com/NVIDIA/cuda-python/releases/download/v13.1.1/cuda-python-v13.1.1.tar.gz Wed, 10 Dec 2025 03:00:34 GMT https://nvidia.github.io/cuda-python/cuda-bindings cuda.core bridges Python's productivity with CUDA's performance through intuitive and pythonic APIs. The mission is to provide users full access to all of the core CUDA features in Python, such as runtime control, compiler and linker. https://nvidia.github.io/cuda-python/cuda-core https://github.com/NVIDIA/cuda-python https://github.com/NVIDIA/cuda-python/releases/download/cuda-core-v0.4.2/cuda-python-cuda-core-v0.4.2.tar.gz Tue, 18 Nov 2025 15:59:04 GMT https://nvidia.github.io/cuda-python/cuda-core The multi-node FFT functionality, available through the cuFFTMp API, enables scientists and engineers to solve distributed 2D and 3D FFTs in exascale problems. The library handles all the communications between machines, allowing users to focus on other aspects of their problems. https://docs.nvidia.com/cuda/cufftmp/ Fri, 07 Nov 2025 19:46:08 GMT https://docs.nvidia.com/cuda/cufftmp/ The multi-node FFT functionality, available through the cuFFTMp API, enables scientists and engineers to solve distributed 2D and 3D FFTs in exascale problems. The library handles all the communications between machines, allowing users to focus on other aspects of their problems. https://docs.nvidia.com/cuda/cufftmp/ Fri, 07 Nov 2025 19:38:26 GMT https://docs.nvidia.com/cuda/cufftmp/ Public API for loading NVIDIA Dynamic Libraries https://nvidia.github.io/cuda-python/cuda-pathfinder/latest/ https://github.com/NVIDIA/cuda-python/tree/main/cuda_pathfinder Mon, 13 Oct 2025 15:43:48 GMT https://nvidia.github.io/cuda-python/cuda-pathfinder/latest/ NVIDIA cuQuantum SDK is a high-performance library for quantum information science and beyond. Currently its primary target is quantum circuit simulations and it contains the major component: - cuStateVec: a high-performance library for state vector computations - cuTensorNet: a high-performance library for tensor network computations License Agreements:- The packages are governed by the NVIDIA cuQuantum Software License Agreement (EULA). By downloading and using the packages, you accept the terms and conditions of the NVIDIA cuQuantum EULA - https://docs.nvidia.com/cuda/cuquantum/license.html https://docs.nvidia.com/cuda/cuquantum/index.html https://developer.nvidia.com/cuquantum-downloads Fri, 05 Sep 2025 04:07:23 GMT https://developer.nvidia.com/cuquantum-sdk CUDA C++ Core Libraries https://docs.nvidia.com/cuda/index.html Fri, 15 Aug 2025 03:05:54 GMT https://developer.nvidia.com/cuda-toolkit Morpheus Cybersecurity Library Thu, 24 Jul 2025 18:47:18 GMT https://github.com/nv-morpheus/Morpheus Morpheus Cybersecurity Library Thu, 24 Jul 2025 18:42:13 GMT https://github.com/nv-morpheus/Morpheus Morpheus Cybersecurity Library Thu, 24 Jul 2025 18:37:25 GMT https://github.com/nv-morpheus/Morpheus A GPU accelerated streaming data library with python bindings Wed, 23 Jul 2025 19:54:57 GMT https://github.com/nv-morpheus/MRC A GPU accelerated streaming data library with python bindings Wed, 23 Jul 2025 19:54:27 GMT https://github.com/nv-morpheus/MRC NVIDIA CUDA Deep Neural Network (cuDNN) is a GPU-accelerated library of primitives for deep neural networks. It provides highly tuned implementations of routines arising frequently in DNN applications. License Agreements:- The packages are governed by the NVIDIA cuDNN Software License Agreement (EULA). By downloading and using the packages, you accept the terms and conditions of the NVIDIA cuDNN EULA - https://docs.nvidia.com/deeplearning/cudnn/sla/index.html https://docs.nvidia.com/deeplearning/cudnn/ https://developer.nvidia.com/rdp/cudnn-download Sat, 12 Jul 2025 20:58:54 GMT https://developer.nvidia.com/cudnn This is a development package for Graph JIT configuration for NVIDIA's cuDNN deep neural network acceleration library Sat, 12 Jul 2025 20:58:37 GMT This is a runtime package for Graph JIT configuration for NVIDIA's cuDNN deep neural network acceleration library. Developers should install libcudnn-jit-dev. Sat, 12 Jul 2025 20:58:03 GMT NVIDIA Holoscan is the AI sensor processing platform that combines hardware systems for low-latency sensor and network connectivity, optimized libraries for data processing and AI, and core microservices to run streaming, imaging, and other applications, from embedded to edge to cloud. It can be used to build streaming AI pipelines for a variety of domains, including Medical Devices, High Performance Computing at the Edge, Industrial Inspection and more. https://docs.nvidia.com/cuda/holoscan/ Thu, 26 Jun 2025 20:10:43 GMT https://docs.nvidia.com/cuda/holoscan/ NVIDIA Holoscan python bindings Thu, 26 Jun 2025 20:07:08 GMT This is a runtime package only. Developers should install holoscan (python) and/or libholoscan-dev (C++ SDK) to build with Holoscan Thu, 26 Jun 2025 20:03:38 GMT NVIDIA CUDA Deep Neural Network (cuDNN) is a GPU-accelerated library of primitives for deep neural networks. It provides highly tuned implementations of routines arising frequently in DNN applications. License Agreements:- The packages are governed by the NVIDIA cuDNN Software License Agreement (EULA). By downloading and using the packages, you accept the terms and conditions of the NVIDIA cuDNN EULA - https://docs.nvidia.com/deeplearning/cudnn/sla/index.html https://docs.nvidia.com/deeplearning/cudnn/ https://developer.nvidia.com/rdp/cudnn-download file:///jenkins/workspace/nvidia/conda/cudnn-builder/files/cudnn-11.0-linux-x64-v8.0.4.30.tgz Sat, 05 Apr 2025 21:28:11 GMT https://developer.nvidia.com/cudnn Use this package to select which ARM variant to use at runtime or compile for. There are a couple of ARM variants that differ based on architecture. Currently the main ones are: * SBSA ( Server Base System Architecture ), which is used in clusters, cloud environments, etc. * Tegra ( System on a Chip ), which is used in edge applications like mobile internet devices, streaming systems, etc. These variants have different CUDA Toolkits, and the arm-variant package is used to select between them. The default is SBSA. Select the tegra 'arm-variant' package by installing 'arm-variant=*=tegra'. Tue, 25 Mar 2025 18:50:09 GMT https://github.com/conda-forge/arm-variant-feedstock NVIDIA cuBLASMp is a high performance, multi-process, GPU accelerated library for distributed basic dense linear algebra. cuBLASMp is compatible with 2D block-cyclic data layout and provides PBLAS-like C APIs. A companion library, CAL, contains utilities to manage communicators and to synchronize processes in a safe way. https://docs.nvidia.com/cuda/cublas/ Tue, 04 Mar 2025 01:42:55 GMT https://docs.nvidia.com/cuda/cublasmp/ NVIDIA cuBLASMp is a high performance, multi-process, GPU accelerated library for distributed basic dense linear algebra. cuBLASMp is compatible with 2D block-cyclic data layout and provides PBLAS-like C APIs. A companion library, CAL, contains utilities to manage communicators and to synchronize processes in a safe way. https://docs.nvidia.com/cuda/cublas/ Tue, 04 Mar 2025 01:42:39 GMT https://docs.nvidia.com/cuda/cublasmp/ This package contains CLI tools for validating and linting NVIDIA's conda recipes on conda-forge. For a description of the tools see the README in the package feedstock. The tools are hosted directly in the feedstock; there is no external source code repository for these tools at this time. Thu, 16 Jan 2025 19:15:43 GMT https://github.com/conda-forge/cf-nvidia-tools-feedstock This is a runtime package only. Developers should install libcudss-dev to build with cuDSS. Mon, 13 Jan 2025 22:53:56 GMT This is a runtime package only. Users should install the corresponding meta-package instead. Developers should install libcudss-dev to build with cuDSS. Mon, 13 Jan 2025 22:53:24 GMT This package is examples only. Developers should install libcudss-dev to build with cuDSS. Mon, 13 Jan 2025 22:52:54 GMT This is a runtime package only. Users should install the corresponding meta-package instead. Developers should install libcudss-dev to build with cuDSS. Mon, 13 Jan 2025 22:52:45 GMT Numba CUDA target https://nvidia.github.io/numba-cuda/ https://github.com/NVIDIA/numba-cuda Mon, 16 Sep 2024 12:53:49 GMT https://nvidia.github.io/numba-cuda/ The cuFFT library provides GPU-accelerated Fast Fourier Transform (FFT) implementations. https://docs.nvidia.com/cuda/cufft/ /jenkins/agent/cia-jenkins-master-03.nvidia.com/workspace/fragments/conda/conda-component-libcufft/conda/build/libcufft-linux-x86_64-10.6.0.54.tar.gz Wed, 14 Aug 2024 19:41:40 GMT https://developer.nvidia.com/cufft CUDA internal headers. https://docs.nvidia.com/cuda/index.html Wed, 14 Aug 2024 19:05:50 GMT https://developer.nvidia.com/cuda-toolkit nvJitLink - Just-in-Time Link Time Optimization (JIT LTO) https://docs.nvidia.com/cuda/nvjitlink/index.html Wed, 14 Aug 2024 19:03:13 GMT https://developer.nvidia.com/cuda-toolkit Compiler for CUDA applications. https://docs.nvidia.com/cuda/index.html Wed, 14 Aug 2024 19:01:23 GMT https://developer.nvidia.com/cuda-toolkit Compiler for CUDA applications. https://docs.nvidia.com/cuda/index.html Wed, 14 Aug 2024 18:59:45 GMT https://developer.nvidia.com/cuda-toolkit