nvidia-cusparse-cu12
CUSPARSE native runtime libraries
Decision gist · record as of 2026-08-14
Yes, if you are installing a higher-level CUDA library (PyTorch, TensorFlow, CuPy) that lists it as a dependency and your system matches the supported platforms (Linux x86_64/aarch64 or Windows x86_64) with CUDA 12 drivers. Do not install directly unless you have a specific reason; it is a low-level runtime library. Verify NVIDIA's proprietary license terms for your use case.AI-flagged interpretation of the facts on this page — verify before relying
Before you install
- Requires NVIDIA CUDA 12 compatible GPU and drivers; only available for Linux (x86_64, aarch64) and Windows (x86_64).
- Medium install friction due to platform-specific wheels (x86_64, aarch64, Windows).
- Package is aging—last release was 435 days ago—but remains actively used in the top 1000 PyPI packages.
License · maintenance · safety
LicenseRef-NVIDIA-Proprietary (unclear) — Licensed under LicenseRef-NVIDIA-Proprietary with unclear treatment. This is a proprietary NVIDIA runtime library; review NVIDIA's licensing terms before deploying in commercial or restricted environments.
last release 2025-06-05 (435 days)
0 known vulnerabilities (OSV.dev, 2026-08-14) · 26,883,888 downloads/mo, #869 on PyPI
Alternatives
Verify before relying
pip install nvidia-cusparse-cu12
import nvidia.cusparse
# Use via higher-level libraries like CuPy or PyTorch that depend on this runtime- Whether this package is meant for direct use or only as a transitive dependency of higher-level CUDA libraries.
- Specific NVIDIA driver version requirements for CUDA 12 compatibility.
- Whether the 435-day age indicates maintenance lag or stable, feature-complete status.
What it is and what it does
nvidia-cusparse-cu12 is a runtime library distribution for NVIDIA's CUSPARSE, a GPU-accelerated library for sparse linear algebra operations. It packages the native binaries needed to perform efficient sparse matrix computations on NVIDIA GPUs using CUDA 12. This is typically not used directly by application code; instead, it serves as a dependency for higher-level Python libraries like CuPy, PyTorch, or TensorFlow that expose sparse tensor operations to users.
The package is platform-specific, with separate wheels for Linux (x86_64 and aarch64) and Windows (x86_64). It requires nvidia-nvjitlink-cu12 as a runtime dependency. Installation has medium friction due to the need to match your system architecture and CUDA environment. The package is in Beta status and has not been updated in 435 days, suggesting it may be feature-stable but not actively developed.
Use it for
- Enable sparse matrix operations in PyTorch or TensorFlow models running on NVIDIA GPUs.
- Support CuPy-based scientific computing workflows requiring efficient sparse linear algebra.
- Provide GPU acceleration for sparse tensor computations in machine learning pipelines.
- Satisfy runtime dependencies when installing CUDA-accelerated Python libraries.
Worth the install?
AI-flagged interpretation of the facts on this page. Verify before relying on it.
Yes, if you are installing a higher-level CUDA library (PyTorch, TensorFlow, CuPy) that lists it as a dependency and your system matches the supported platforms (Linux x86_64/aarch64 or Windows x86_64) with CUDA 12 drivers.
Do not install directly unless you have a specific reason; it is a low-level runtime library. Verify NVIDIA's proprietary license terms for your use case.
Install
nvidia-cusparse-cu12 on PyPI
Before you install
Medium install friction due to platform-specific wheels (x86_64, aarch64, Windows). Package is aging—last release was 435 days ago—but remains actively used in the top 1000 PyPI packages. Depends on nvidia-nvjitlink-cu12.
Requires NVIDIA CUDA 12 compatible GPU and drivers; only available for Linux (x86_64, aarch64) and Windows (x86_64).
License in practice
Licensed under LicenseRef-NVIDIA-Proprietary with unclear treatment. This is a proprietary NVIDIA runtime library; review NVIDIA's licensing terms before deploying in commercial or restricted environments.
Quickstart
pip install nvidia-cusparse-cu12
import nvidia.cusparse
# Use via higher-level libraries like CuPy or PyTorch that depend on this runtimeVerify before relying
- Whether this package is meant for direct use or only as a transitive dependency of higher-level CUDA libraries.
- Specific NVIDIA driver version requirements for CUDA 12 compatibility.
- Whether the 435-day age indicates maintenance lag or stable, feature-complete status.
Package facts
| License | LicenseRef-NVIDIA-Proprietary unclear |
| Python support | Supports the current Python release >=3 |
| Install friction | Medium. Platform-specific wheel |
| Runtime dependencies | 1 packagenvidia-nvjitlink-cu12 |
| Maintenance | Aging 435 days since the last release |
| First released | |
| Downloads | 26,883,888 / month, #869 on PyPI 30-day window, as of 2026-08-14 |
| Known vulnerabilities | None known OSV.dev, checked 2026-08-14 |
| Classifiers | Development Status :: 4 - BetaIntended Audience :: DevelopersIntended Audience :: EducationIntended Audience :: Science/ResearchLicense :: Other/Proprietary LicenseNatural Language :: EnglishOperating System :: Microsoft :: WindowsOperating System :: POSIX :: LinuxProgramming Language :: Python :: 3Programming Language :: Python :: 3 :: OnlyProgramming Language :: Python :: 3.10Programming Language :: Python :: 3.11Programming Language :: Python :: 3.5Programming Language :: Python :: 3.6Programming Language :: Python :: 3.7Programming Language :: Python :: 3.8Programming Language :: Python :: 3.9Topic :: Scientific/EngineeringTopic :: Scientific/Engineering :: Artificial IntelligenceTopic :: Scientific/Engineering :: MathematicsTopic :: Software DevelopmentTopic :: Software Development :: Libraries |
Evidence: nvidia_cusparse_cu12-12.5.10.65-py3-none-manylinux2014_aarch64.manylinux_2_17_aarch64.whl; nvidia_cusparse_cu12-12.5.10.65-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl; nvidia_cusparse_cu12-12.5.10.65-py3-none-win_amd64.whl
Tags
Let your AI agent find packages like this
Example. Real query, live index.
You found this page by searching. An agent finds it by wishing: SkillFed indexes 14,416 PyPI packages by what they can do, searchable in plain language.
wish › “nvidia cusparse gpu sparse matrix”
- nvidia-cusparse-cu12Provides NVIDIA CUSPARSE native runtime libraries for GPU-accelerated…
- nvidia-cusparseProvides NVIDIA CUSPARSE native runtime libraries for GPU-accelerated…
- nvidia-cusparse-cu11Provides CUSPARSE native runtime libraries for CUDA 11, enabling…
Give your agent the search over MCP, or paste the wish link into any chat.
More Software Development packages
Provides backported and experimental type hints for Python 3.9+, allowing use of newer typing features on older Python versions and enabling early experimentation with type system PEPs before they enter the standard library.
NumPy provides an N-dimensional array object and a comprehensive suite of mathematical, linear algebra, Fourier transform, and random number functions for scientific computing in Python.
FastAPI is a Python web framework for building REST APIs using type hints, with automatic request validation, serialization, and interactive API documentation.
Provides a way to document function parameters, class attributes, return types, and variables inline using Python's `Annotated` type hint syntax instead of traditional docstrings.
Typer builds command-line applications from Python functions using type hints, automatically generating help text, argument parsing, and shell completion.
Install it if you are building CLIs in Python.
Distlib provides low-level packaging utilities for building, distributing, and managing Python software—including metadata handling, version specifiers, wheel support, script installation, and dependency resolution.
See also nvidia-cusparse · nvidia-cusparse-cu11 · cuda-toolkit · nvidia-libnvcomp-cu12 · nvidia-cusolver-cu12 · nvidia-cublas-cu12 · nvidia-mathdx · nvidia-cusparselt-cu13 · nvidia-cusolver-cu11 · nvidia-cusolver