nvidia-cudnn-cu12
cuDNN runtime libraries
Install
nvidia-cudnn-cu12 on PyPI
pip
pip install nvidia-cudnn-cu12uv
uv add nvidia-cudnn-cu12poetry
poetry add nvidia-cudnn-cu12Package facts
| License | not declared (unclear) |
| Python support | supports the current Python release (>=3) |
| Install friction | medium — platform-specific wheel |
| Runtime dependencies | 1 — nvidia-cublas-cu12 |
| Maintenance | actively maintained — 42 days since the last release |
| 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: nvidia_cudnn_cu12-9.24.0.43-py3-none-manylinux_2_27_aarch64.whl; nvidia_cudnn_cu12-9.24.0.43-py3-none-manylinux_2_27_x86_64.whl; nvidia_cudnn_cu12-9.24.0.43-py3-none-win_amd64.whl
Keywords: cuda, nvidia, runtime, machine learning, deep learning
About nvidia-cudnn-cu12
from the package's own PyPI description — quoted content, verbatim
cuDNN runtime libraries containing primitives for deep neural networks.
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 cuDNN runtime libraries—optimized primitives for deep neural network operations—packaged for CUDA 12 environments on Linux, Windows, and ARM64 systems.
Medium install friction due to platform-specific wheel distributions (Linux x86_64, ARM64, Windows) and a single compiled runtime dependency on nvidia-cublas-cu12. Package is actively maintained with a recent release 42 days ago.
License treatment is 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-cudnn-cu12==9.24.0.43
import nvidia.cudnn
Requires CUDA 12 toolkit and nvidia-cublas-cu12 installed; only runs on Linux (x86_64/ARM64) or Windows; Python ≥3.0 required.
Verdict: A well-maintained, actively-developed cuDNN runtime for CUDA 12 in the top 1000 PyPI packages with zero known vulnerabilities. Installation requires platform-specific wheels and compiled GPU dependencies. The unclear license status is a notable gap for compliance-conscious users; clarification from NVIDIA is advisable before production deployment.
Needs verification
- What is the actual license governing this package? (SPDX or full text not provided)
- Does this package require NVIDIA GPU hardware to function, or is it usable on CPU-only systems?
- Are there version compatibility constraints between nvidia-cudnn-cu12 and nvidia-cublas-cu12 beyond pip's dependency resolution?
Similar packages
unclear · top 1,000 on PyPI
torchpermissive · top 1,000 on PyPI
onnxruntimepermissive · top 1,000 on PyPI
nvidia-cuda-runtimeunclear · top 1,000 on PyPI
nvidia-curand-cu12unclear · top 1,000 on PyPI
tritonpermissive · top 1,000 on PyPI
nvidia-cuda-runtime-cu12unclear · top 1,000 on PyPI
nvidia-cublas-cu12unclear · top 1,000 on PyPI
nvidia-cuda-cupti-cu12unclear · top 1,000 on PyPI
nvidia-nccl-cu13unclear · top 1,000 on PyPI