All skills
holoscan-install-wheel
Install Holoscan SDK 5.x CUDA 13 Python wheels in a fresh venv and verify the import.
- Part of NVIDIA/skills
- 1.8k installs
- Hardware & IoT · Embedded development
When to use it
You have a Linux dGPU host with CUDA 13 and want Holoscan 5.x Python bindings, but need to confirm platform prerequisites and pick full-stack or component wheels.
For example
You say
Use holoscan-install-wheel to install Holoscan SDK 5.0 in a fresh venv on my Ubuntu 24.04 x86_64 machine with a CUDA 13 driver, then verify the import.
You get
- Pinned wheel install
- Import and version check
- Version report
How it works
- Checks host, Python, glibc, and driver before installing, and stops on a failed prerequisite.
- Offers a source build instead when the requested 5.x wheel is not published.
- Runs the first_pipeline example for runtime coverage, not just an import.
Before installing
Requires regular CPython 3.12–3.14, glibc 2.39+, x86_64 or arm64, and a CUDA 13 driver; no CUDA 12 or Tegra/iGPU variant exists.
More Embedded development skills
See allAlso in this pack
- accelerated-computing-cudf3.2kGuides GPU DataFrame work with cuDF, including pandas acceleration, joins, groupby, file I/O and multi-GPU jobs.
- cuopt-developer2.8kModifies, builds, tests, and debugs NVIDIA cuOpt source code and prepares contributions.
- cuopt-install2.6kInstalls cuOpt for Python, C, or server via pip, conda, or Docker and verifies it works.
- data-designer2.6kGenerates a synthetic dataset from your description of the data you need.