nvidia-cufile

cuFile GPUDirect libraries

nvidia-cufile License unclear Active v1.18.1.6 released 2026-06-29

Install

nvidia-cufile on PyPI

pip

pip install nvidia-cufile

uv

uv add nvidia-cufile

poetry

poetry add nvidia-cufile

Package facts

License	not declared (unclear)
Python support	supports the current Python release (>=3)
Install friction	medium — platform-specific wheel
Runtime dependencies	none
Maintenance	actively maintained — 45 days since the last release
First released	2025-04-04
Popularity	one of the top 1,000 most-downloaded packages on PyPI (30-day window, as of 2026-08-13)
Known vulnerabilities	none known (OSV.dev, checked 2026-08-13)

Evidence: nvidia_cufile-1.18.1.6-py3-none-manylinux_2_27_aarch64.whl; nvidia_cufile-1.18.1.6-py3-none-manylinux_2_27_x86_64.whl

Keywords: cuda, nvidia, runtime, machine learning, deep learning

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

About nvidia-cufile

from the package's own PyPI description — quoted content, verbatim

cuFile GPUDirect libraries

Read as markdown · JSON record · Homepage

AI interpretation — verify before relying

AI-generated interpretation of the package facts above; every digit, version, license, or vulnerability id it cites is grounded in the facts already shown on this page

Provides NVIDIA cuFile GPUDirect libraries for accelerated GPU-direct file I/O operations, enabling direct data transfers between storage and GPU memory without CPU involvement.

Medium install friction due to platform-specific wheels (x86_64 and aarch64 Linux only); actively maintained with a release 45 days ago. No runtime dependencies simplifies deployment once the wheel matches your architecture.

License terms are unclear—no SPDX identifier or raw license text is available in the package metadata, making it difficult to assess compliance obligations before use.

Usage

pip install nvidia-cufile==1.18.1.6

Platform-specific wheels available only for Linux x86_64 and aarch64 architectures; Windows classifier present but no corresponding wheel provided.

Verdict: A specialized NVIDIA library for GPU-accelerated I/O in machine learning and scientific computing workflows. Active maintenance and no known vulnerabilities are positive signals, but unclear licensing and platform-specific wheels (Linux only) narrow its audience. Best suited for teams already committed to NVIDIA CUDA infrastructure.

Needs verification

What the actual license terms are (NVIDIA proprietary, BSD, or other) given the metadata gap.
Whether Windows support (listed in classifiers) is actually available given only Linux wheels are present.
What system-level dependencies or NVIDIA CUDA versions are required for this package to function.

gpu direct file i/onvidia cufilecuda gpu storage accessgpudirect storagenvidia cuda file operationsgpu memory file transfercuda direct i/o