pyvista-zstd

What it is and what it does

pyvista-zstd is a compression library that writes VTK datasets to a custom .pv format using Zstandard (zstd) compression instead of VTK's built-in XML writer. It wraps PyVista and zstandard to provide faster I/O with better compression ratios. The library automatically registers with PyVista's reader system, so once installed, pv.read() handles .pv files directly without additional imports.

The main advantage over VTK's XML format is speed and compression efficiency. According to the description, it achieves significantly higher write throughput (up to 1845 MB/s with threading) and better compression ratios (3.02 vs 2.52 for zlib) while using less disk space overall. It supports multi-threaded compression and decompression, and optimizes storage for fixed-width cell topologies by storing the cell width once instead of repeating it for every cell.

Use it for:

• Save large VTK meshes to disk quickly when working with scientific simulations or 3D geometry data.
• Reduce storage footprint for archived VTK datasets while maintaining fast read access.
• Parallelize compression of multiple VTK files in batch workflows using the multi-threaded write API.
• Integrate compressed mesh I/O into PyVista-based visualization or analysis pipelines without code changes.
• Replace VTK XML format in existing projects where write speed or compression ratio is a bottleneck.

Worth the install?

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

Compress and decompress VTK datasets using Zstandard compression with multi-threaded read and write operations, integrated with PyVista.

Yes, if you work with VTK datasets in PyVista and care about I/O performance or file size. The package is actively maintained, has no known vulnerabilities, and integrates seamlessly with PyVista. Install friction is low. The MIT license poses no restrictions. Best suited for workflows where mesh file I/O is a performance concern; less critical for one-off analysis or small datasets.

Install

pyvista-zstd on PyPI

pip install pyvista-zstd

uv add pyvista-zstd

poetry add pyvista-zstd

Installing pyvista-zstd

Before you install

Low install friction with a pure-Python wheel. Active maintenance: released 4 days ago with last commit on 2026-08-10. Supports Python 3.10–3.13. Three runtime dependencies (pyvista, tqdm, zstandard) are all stable, widely-used libraries.

License in practice

MIT license is permissive; you can use this in commercial and proprietary projects with minimal restrictions.

Quickstart

pip install pyvista-zstd

import pyvista_zstd
import pyvista as pv

ds = pv.Sphere()
pyvista_zstd.write(ds, "dataset.pv")
ds_in = pyvista_zstd.read("dataset.pv")

Verify before relying

• Whether the 37x read speedup and 14x write speedup claims apply to typical real-world datasets or only to the specific benchmark case shown in the description.
• Compatibility with VTK dataset types beyond those explicitly mentioned (triangles, quads, tetrahedra).
• Whether the .pv format is stable across future versions or if there are breaking changes to expect.

Package facts

Evidence: pyvista_zstd-0.3.0-py3-none-any.whl

Tags

More Scientific/Engineering packages