nvidia-cuda-nvrtc
NVRTC native runtime libraries
Decision gist · record as of 2026-08-14
Yes, if you need runtime CUDA kernel compilation and are working on a supported platform (Linux x86_64/aarch64 or Windows x86_64). The package is actively maintained, has no known vulnerabilities, and is widely used (top 1000 PyPI by downloads). The main caveat is the unclear license—verify the license terms before using in proprietary or redistributed work. Medium install friction is acceptable for a platform-specific native library.AI-flagged interpretation of the facts on this page — verify before relying
Before you install
- Requires a compatible CUDA-capable GPU and NVIDIA driver; wheels are platform-specific (Linux x86_64, Linux aarch64, or Windows x86_64 only).
- Medium install friction due to platform-specific wheel distributions (manylinux2010/2014 x86_64, aarch64, Windows).
- Package is actively maintained with a release within the last 80 days.
License · maintenance · safety
(unclear) — License treatment is unclear—no SPDX identifier or raw license text is declared in the package metadata, so the legal terms governing use and redistribution are not explicitly stated in the package itself.
last release 2026-05-26 (80 days)
0 known vulnerabilities (OSV.dev, 2026-08-14) · 43,231,966 downloads/mo, #645 on PyPI
Alternatives
Verify before relying
pip install nvidia-cuda-nvrtc==13.3.33
import nvidia.cuda.nvrtc as nvrtc
# Use NVRTC API for runtime kernel compilation- Whether this package requires a system-level CUDA installation or provides self-contained NVRTC binaries.
- Exact license terms and any redistribution restrictions not declared in package metadata.
- Whether the package supports Python 3.5 and 3.6 as classifiers claim, given modern CUDA toolchain constraints.
What it is and what it does
nvidia-cuda-nvrtc is a Python wrapper around NVIDIA's NVRTC (NVIDIA Runtime Compilation) native libraries, which allow you to compile CUDA C++ kernel code at runtime rather than at build time. This is useful when you need to generate or modify GPU kernels dynamically based on runtime conditions—for example, in machine learning frameworks that auto-tune kernel implementations or in scientific computing pipelines that specialize kernels for specific problem sizes.
The package ships as platform-specific wheels for Linux (x86_64 and aarch64) and Windows (x86_64), with no additional Python runtime dependencies. It targets developers and researchers working with CUDA, machine learning, and scientific computing. The package is actively maintained and supports Python 3 across multiple versions, though installation requires a compatible NVIDIA GPU and driver.
Use it for
- Dynamically compile CUDA kernels in machine learning frameworks that need to optimize code for different hardware or problem parameters at runtime.
- Build scientific computing applications that generate specialized GPU kernels on-the-fly for numerical algorithms.
- Develop GPU-accelerated data processing pipelines where kernel specialization improves performance without pre-compilation.
- Create educational tools or research prototypes that experiment with CUDA kernel generation and optimization.
- Integrate runtime kernel compilation into heterogeneous compute workflows that mix CPU and GPU code paths.
Worth the install?
AI-flagged interpretation of the facts on this page. Verify before relying on it.
Yes, if you need runtime CUDA kernel compilation and are working on a supported platform (Linux x86_64/aarch64 or Windows x86_64).
The package is actively maintained, has no known vulnerabilities, and is widely used (top 1000 PyPI by downloads). The main caveat is the unclear license—verify the license terms before using in proprietary or redistributed work. Medium install friction is acceptable for a platform-specific native library.
Install
nvidia-cuda-nvrtc on PyPI
Before you install
Medium install friction due to platform-specific wheel distributions (manylinux2010/2014 x86_64, aarch64, Windows). Package is actively maintained with a release within the last 80 days. No runtime dependencies to manage.
Requires a compatible CUDA-capable GPU and NVIDIA driver; wheels are platform-specific (Linux x86_64, Linux aarch64, or Windows x86_64 only).
License in practice
License treatment is unclear—no SPDX identifier or raw license text is declared in the package metadata, so the legal terms governing use and redistribution are not explicitly stated in the package itself.
Quickstart
pip install nvidia-cuda-nvrtc==13.3.33
import nvidia.cuda.nvrtc as nvrtc
# Use NVRTC API for runtime kernel compilationVerify before relying
- Whether this package requires a system-level CUDA installation or provides self-contained NVRTC binaries.
- Exact license terms and any redistribution restrictions not declared in package metadata.
- Whether the package supports Python 3.5 and 3.6 as classifiers claim, given modern CUDA toolchain constraints.
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 80 days since the last release |
| First released | |
| Downloads | 43,231,966 / month, #645 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_cuda_nvrtc-13.3.33-py3-none-manylinux2010_x86_64.manylinux_2_12_x86_64.whl; nvidia_cuda_nvrtc-13.3.33-py3-none-manylinux2014_aarch64.manylinux_2_17_aarch64.whl; nvidia_cuda_nvrtc-13.3.33-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 › “gpu kernel runtime compile”
- nvidia-cuda-nvrtcProvides NVIDIA CUDA NVRTC (NVIDIA Runtime Compilation) native…
- nvidia-cuda-nvccProvides the NVIDIA CUDA nvcc compiler for building CUDA applications…
- cuda-pythoncuda-python is a metapackage providing Pythonic access to NVIDIA's…
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-cuda-nvrtc-cu11 · nvidia-cuda-nvrtc-cu12 · nvidia-cuda-nvcc · nvidia-nvvm · nvidia-cuda-nvcc-cu12 · nvidia-cuda-tileiras · nvidia-nvfatbin · nvidia-nvjitlink · nvidia-curand-cu11 · nvidia-cuda-runtime-cu12