nvidia-cufft-cu11

What it is and what it does

nvidia-cufft-cu11 is a binary distribution of NVIDIA's CUFFT library—a GPU-accelerated Fast Fourier Transform runtime—packaged for Python environments targeting CUDA 11. It provides no Python API of its own; instead, it supplies the native libraries that higher-level packages depend on to perform FFT operations on NVIDIA GPUs. The package is platform-specific, with separate wheels for Linux (x86_64 and aarch64) and Windows.

The package has been abandoned with no active repository or maintenance. It carries an unclear proprietary license from NVIDIA and no documented security or bug-fix support. Users installing this package are typically doing so as a transitive dependency of a machine learning or scientific computing library, rather than directly. For new projects, checking whether a newer CUDA version or alternative distribution channel is available is advisable.

Use it for:

• Satisfy CUFFT runtime dependencies for GPU-accelerated libraries that require CUDA 11.
• Enable GPU-accelerated FFT computations in signal processing or scientific computing workflows on NVIDIA hardware.
• Support legacy machine learning pipelines built against CUDA 11 that cannot migrate to newer CUDA versions.
• Provide native FFT libraries for containerized or isolated Python environments targeting NVIDIA GPUs.

Worth the install?

AI-flagged interpretation of the facts on this page — verify before relying

Provides NVIDIA CUFFT native runtime libraries for CUDA 11, enabling GPU-accelerated Fast Fourier Transform computations on compatible systems.

Yes, but with caution. Install only if required as a transitive dependency of an active package, or if you are locked to CUDA 11 and cannot upgrade. The package is abandoned and carries an unclear proprietary license; verify NVIDIA's current licensing terms and consider whether a newer CUDA version is available for your use case. No known vulnerabilities, but no active maintenance either.

Install

nvidia-cufft-cu11 on PyPI

pip install nvidia-cufft-cu11

uv add nvidia-cufft-cu11

poetry add nvidia-cufft-cu11

Installing nvidia-cufft-cu11

Before you install

Medium install friction due to platform-specific wheels for x86_64 and aarch64 Linux, Windows, and legacy manylinux1. Package is abandoned with no active maintenance or repository activity visible.

License in practice

Licensed under NVIDIA Proprietary Software with unclear treatment. Users should verify licensing terms directly with NVIDIA before deployment, particularly in commercial or redistributed contexts.

Quickstart

pip install nvidia-cufft-cu11==10.9.0.58

import nvidia.cufft
# Use CUFFT runtime for FFT operations via CUDA-enabled libraries

Requires NVIDIA CUDA 11 toolkit and compatible GPU; only available for Linux (x86_64, aarch64) and Windows (x86_64).

Verify before relying

• Whether this package is still compatible with current CUDA 11.x minor versions and modern GPU architectures.
• Whether NVIDIA Proprietary Software license permits use in open-source or commercial projects without additional agreements.
• Whether abandonment means security updates or bug fixes are no longer available.
• Whether newer CUDA versions have superseded this package with different naming or distribution.

Package facts

Evidence: nvidia_cufft_cu11-10.9.0.58-py3-none-manylinux1_x86_64.whl; nvidia_cufft_cu11-10.9.0.58-py3-none-manylinux2014_aarch64.whl; nvidia_cufft_cu11-10.9.0.58-py3-none-manylinux2014_x86_64.whl; nvidia_cufft_cu11-10.9.0.58-py3-none-win_amd64.whl

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

Tags

More Software Development packages

Further reading