nvidia-cublas-cu12
CUBLAS native runtime libraries
Decision gist · record as of 2026-08-14
Yes, if you are running GPU-accelerated machine learning or scientific computing on NVIDIA hardware with CUDA 12. This is a foundational dependency pulled in by frameworks like PyTorch or TensorFlow, so you typically install it indirectly. Install it directly only if you are building a custom CUDA application or troubleshooting a missing runtime dependency. No, if you do not have NVIDIA GPU hardware or are not using CUDA 12—it will not install or function on incompatible systems.AI-flagged interpretation of the facts on this page — verify before relying
Before you install
- Requires NVIDIA GPU hardware and CUDA 12 toolkit compatibility; platform-specific wheels limit installation to x86_64, aarch64 Linux, or Windows amd64.
- Medium install friction due to platform-specific wheels (x86_64, aarch64, Windows).
- Active maintenance as of 128 days ago.
License · maintenance · safety
LicenseRef-NVIDIA-Proprietary (unclear) — Licensed under LicenseRef-NVIDIA-Proprietary with unclear treatment. This is NVIDIA's proprietary software; review NVIDIA's licensing terms before deploying in commercial or restricted environments.
last release 2026-04-08 (128 days)
0 known vulnerabilities (OSV.dev, 2026-08-14) · 27,569,652 downloads/mo, #856 on PyPI
Alternatives
Verify before relying
pip install nvidia-cublas-cu12
import nvidia.cublas.cu12- Whether this package alone is sufficient or if additional CUDA toolkit components must be installed separately
- Specific GPU models or compute capabilities required for full functionality
- Whether the package works on non-standard platforms or requires specific system libraries
What it is and what it does
nvidia-cublas-cu12 is a distribution of NVIDIA's CUBLAS (CUDA Basic Linear Algebra Subroutines) native runtime libraries compiled for CUDA 12. It provides the low-level GPU-accelerated linear algebra primitives that machine learning frameworks and scientific computing libraries depend on to perform matrix operations, tensor computations, and other mathematical workloads on NVIDIA GPUs.
This package is not typically imported directly by end users; instead, it serves as a runtime dependency for higher-level frameworks like PyTorch, TensorFlow, or other CUDA-aware libraries. Installation requires an NVIDIA GPU and CUDA 12 compatibility. The package is distributed as platform-specific wheels for Linux (x86_64 and aarch64) and Windows (amd64), and depends on nvidia-cuda-nvrtc-cu12 for compilation support.
Use it for
- Enable GPU acceleration in PyTorch or TensorFlow models by providing the underlying CUBLAS runtime they depend on
- Run machine learning inference or training workloads that require fast matrix multiplication on NVIDIA GPUs
- Support scientific computing applications (NumPy-based, JAX, CuPy) that offload linear algebra to GPU
- Deploy containerized deep learning applications that need CUDA 12 math libraries without system-wide CUDA installation
Worth the install?
AI-flagged interpretation of the facts on this page. Verify before relying on it.
Yes, if you are running GPU-accelerated machine learning or scientific computing on NVIDIA hardware with CUDA 12.
This is a foundational dependency pulled in by frameworks like PyTorch or TensorFlow, so you typically install it indirectly. Install it directly only if you are building a custom CUDA application or troubleshooting a missing runtime dependency. No, if you do not have NVIDIA GPU hardware or are not using CUDA 12—it will not install or function on incompatible systems.
Install
nvidia-cublas-cu12 on PyPI
Before you install
Medium install friction due to platform-specific wheels (x86_64, aarch64, Windows). Active maintenance as of 128 days ago. Requires nvidia-cuda-nvrtc-cu12 as a runtime dependency, indicating tight coupling to NVIDIA's CUDA ecosystem.
Requires NVIDIA GPU hardware and CUDA 12 toolkit compatibility; platform-specific wheels limit installation to x86_64, aarch64 Linux, or Windows amd64.
License in practice
Licensed under LicenseRef-NVIDIA-Proprietary with unclear treatment. This is NVIDIA's proprietary software; review NVIDIA's licensing terms before deploying in commercial or restricted environments.
Quickstart
pip install nvidia-cublas-cu12
import nvidia.cublas.cu12Verify before relying
- Whether this package alone is sufficient or if additional CUDA toolkit components must be installed separately
- Specific GPU models or compute capabilities required for full functionality
- Whether the package works on non-standard platforms or requires specific system libraries
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-cuda-nvrtc-cu12 |
| Maintenance | Actively maintained 128 days since the last release |
| First released | |
| Downloads | 27,569,652 / month, #856 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/ResearchNatural 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_cublas_cu12-12.9.2.10-py3-none-manylinux_2_27_aarch64.whl; nvidia_cublas_cu12-12.9.2.10-py3-none-manylinux_2_27_x86_64.whl; nvidia_cublas_cu12-12.9.2.10-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 cublas cuda 12 runtime”
- nvidia-cublas-cu12Provides NVIDIA CUBLAS native runtime libraries for CUDA 12, enabling…
- nvidia-cublas-cu11Provides NVIDIA CUBLAS native runtime libraries for CUDA 11, enabling…
- nvidia-cublasProvides NVIDIA CUBLAS native runtime libraries for GPU-accelerated…
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-cublas · nvidia-cublas-cu11 · blis · nvidia-cusparse-cu12 · nvidia-cusolver · nvidia-cusolver-cu12 · nvidia-cusolver-cu11 · nvidia-cusparse · linear-operator · nvidia-cufft-cu12