cuda-bindings
Python bindings for CUDA
Install
cuda-bindings on PyPI
pip
pip install cuda-bindingsuv
uv add cuda-bindingspoetry
poetry add cuda-bindingsPackage facts
| License | LicenseRef-NVIDIA-SOFTWARE-LICENSE (unclear) |
| Python support | supports the current Python release (>=3.10) |
| Install friction | medium — platform-specific wheel |
| Runtime dependencies | 1 — cuda-pathfinder |
| Maintenance | actively maintained — 76 days since the last release |
| Last repo commit | |
| First released | |
| 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: cuda_bindings-13.3.1-cp310-cp310-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl; cuda_bindings-13.3.1-cp310-cp310-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl; cuda_bindings-13.3.1-cp310-cp310-win_amd64.whl; cuda_bindings-13.3.1-cp311-cp311-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl; cuda_bindings-13.3.1-cp311-cp311-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl; cuda_bindings-13.3.1-cp311-cp311-win_amd64.whl; cuda_bindings-13.3.1-cp312-cp312-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl; cuda_bindings-13.3.1-cp312-cp312-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl; cuda_bindings-13.3.1-cp312-cp312-win_amd64.whl; cuda_bindings-13.3.1-cp313-cp313-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl; cuda_bindings-13.3.1-cp313-cp313-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl; cuda_bindings-13.3.1-cp313-cp313-win_amd64.whl; cuda_bindings-13.3.1-cp314-cp314-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl; cuda_bindings-13.3.1-cp314-cp314-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl; cuda_bindings-13.3.1-cp314-cp314t-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl; cuda_bindings-13.3.1-cp314-cp314t-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl; cuda_bindings-13.3.1-cp314-cp314t-win_amd64.whl; cuda_bindings-13.3.1-cp314-cp314-win_amd64.whl
About cuda-bindings
from the package's own PyPI description — quoted content, verbatim
.. SPDX-FileCopyrightText: Copyright (c) 2025 NVIDIA CORPORATION & AFFILIATES. All rights reserved. .. SPDX-License-Identifier: LicenseRef-NVIDIA-SOFTWARE-LICENSE
cuda-bindings: Low-level CUDA interfaces
.. image:: https://img.shields.io/badge/NVIDIA-black?logo=nvidia :target: https://www.nvidia.com/ :alt: NVIDIA
cuda.bindings <https://nvidia.github.io/cuda-python/cuda-bindings/> is a standard set of low-level interfaces, providing full coverage of and 1:1 access to the CUDA host APIs from Python. Checkout the Overview <https://nvidia.github.io/cuda-python/cuda-bindings/latest/overview.html> for the workflow and performance results.
Repository <https://github.com/NVIDIA/cuda-python/tree/main/cuda_bindings>_Documentation <https://nvidia.github.io/cuda-python/cuda-bindings/>_Examples <https://github.com/NVIDIA/cuda-python/tree/main/cuda_bindings/examples>_Issue tracker <https://github.com/NVIDIA/cuda-python/issues/>_
For the installation instruction, please refer to the Installation <https://nvidia.github.io/cuda-python/cuda-bindings/latest/install.html>_ page.
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 low-level Python bindings for CUDA host APIs, enabling direct 1:1 access to NVIDIA's CUDA runtime from Python with full API coverage.
Medium install friction due to platform-specific wheels (Linux x86_64/aarch64, Windows) and a runtime dependency on cuda-pathfinder. Actively maintained with recent releases.
Licensed under LicenseRef-NVIDIA-SOFTWARE-LICENSE, a non-standard NVIDIA proprietary license. The unclear treatment means you should review NVIDIA's actual license terms before use in commercial or redistributed contexts.
Usage
pip install cuda-bindings
import cuda.bindings
# Access CUDA host APIs directly via cuda.bindings module
Requires Python ≥3.10 and a compatible platform (Linux x86_64/aarch64 or Windows x86_64); cuda-pathfinder must be installed as a runtime dependency.
Verdict: A well-maintained, actively developed package from NVIDIA offering direct CUDA API access for Python developers targeting GPU compute. No known vulnerabilities and broad Python version support (3.10–3.14) are strong signals, but the proprietary license and medium install friction warrant careful evaluation of licensing terms and platform compatibility before adoption.
Needs verification
- Whether LicenseRef-NVIDIA-SOFTWARE-LICENSE permits commercial use, modification, or redistribution without additional agreements.
- Whether cuda-pathfinder is freely available and what its own licensing or system requirements are.
- Performance overhead or compatibility caveats when using these bindings versus direct CUDA C calls.
Similar packages
permissive · top 1,000 on PyPI
dulwichunclear · top 1,000 on PyPI
nvidia-cublas-cu12unclear · top 1,000 on PyPI
nvidia-cufft-cu12unclear · top 1,000 on PyPI
nvidia-cublasunclear · top 1,000 on PyPI
nvidia-cuda-nvrtcunclear · top 1,000 on PyPI
nvidia-cuda-nvrtc-cu12unclear · top 1,000 on PyPI
nvidia-cusparse-cu12unclear · top 1,000 on PyPI
nvidia-cufileunclear · top 1,000 on PyPI
cuda-toolkitunclear · top 1,000 on PyPI