nvidia-cuda-nvcc

What it is and what it does

nvidia-cuda-nvcc is a Python package that bundles the NVIDIA CUDA nvcc compiler, enabling developers to compile CUDA C/C++ code as part of Python-based build workflows. It packages the compiler with three runtime dependencies (nvidia-nvvm, nvidia-cuda-runtime, nvidia-cuda-crt) to provide a self-contained compilation environment. The package is classified as Beta and targets developers, educators, and researchers working with GPU-accelerated computing, machine learning, and deep learning applications.

Installation requires a CUDA-capable NVIDIA GPU and the appropriate platform (Linux aarch64, Linux x86_64, or Windows). The package supports Python 3.5 through 3.11 and is actively maintained, with its latest release dated 2026-06-29. It carries no known security vulnerabilities as of the last check.

Use it for:

• Compile custom CUDA kernels as part of a Python machine learning pipeline or build system.
• Develop GPU-accelerated libraries that require CUDA source compilation during installation.
• Integrate CUDA compilation into continuous integration workflows for GPU-accelerated applications.
• Build custom CUDA extensions for deep learning frameworks that need kernel-level optimization.

Worth the install?

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

Provides the NVIDIA CUDA nvcc compiler for building CUDA applications from Python, bundling the compiler with its runtime dependencies.

Yes, if you need to compile CUDA code from Python on a GPU-equipped system. The package is actively maintained, has no known vulnerabilities, and is widely used (top 5000 on PyPI). Install friction is moderate due to platform-specific wheels and GPU hardware requirements, but this is inherent to CUDA development. Verify that your system meets the GPU and driver prerequisites before installing.

Install

nvidia-cuda-nvcc on PyPI

pip install nvidia-cuda-nvcc

uv add nvidia-cuda-nvcc

poetry add nvidia-cuda-nvcc

Installing nvidia-cuda-nvcc

Before you install

Medium install friction due to platform-specific wheels (Linux aarch64, Linux x86_64, Windows) and three NVIDIA runtime dependencies (nvidia-nvvm, nvidia-cuda-runtime, nvidia-cuda-crt). Active maintenance with a recent release.

Quickstart

pip install nvidia-cuda-nvcc

import nvidia.cuda.nvcc
# Use nvcc compiler for CUDA source files

Requires NVIDIA GPU hardware and CUDA-capable system; wheels are platform-specific (Linux or Windows only).

Verify before relying

• Whether the package provides direct CLI access to nvcc or only programmatic bindings.
• What CUDA compute capabilities and driver versions are supported by version 13.3.73.
• Whether documentation or examples are available beyond the NVIDIA developer homepage.

Package facts

Evidence: nvidia_cuda_nvcc-13.3.73-py3-none-manylinux2014_aarch64.manylinux_2_17_aarch64.whl; nvidia_cuda_nvcc-13.3.73-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl; nvidia_cuda_nvcc-13.3.73-py3-none-win_amd64.whl

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

Tags

More Software Development packages

Further reading