cvista core: VTK Common/Filters/Imaging compute kernels (rendering-free, IO-free, offline).
Project Links
Meta
Requires Python: >=3.11
Classifiers
cvista
cvista is a fast, lean fork of VTK 9.6.2, maintained by the PyVista community as a drop-in graphics layer for PyVista. It is byte-for-byte identical to stock VTK 9.6.2 by default, BSD-3 licensed, and developed in the open at pyvista/cvista. It is not affiliated with Kitware.
cvista is published as three tiers that share the cvista import package:
cvistacore: VTK Common/Filters/Imaging compute kernels (rendering-free, IO-free)cvista-rendering: the OpenGL2/FreeType/Charts/Views rendering stackcvista-io: every VTK reader and writer
This package: cvista
The base tier. It provides VTK's Common, Filters, and Imaging modules: the data model and compute kernels, with no rendering and no data IO. Install it alone for offline geometry and array processing, or opt into the other tiers with the rendering, io, and all extras.
pip install cvista # core only
pip install cvista[rendering] # + rendering tier
pip install cvista[io] # + data IO tier
pip install cvista[all] # everything
Links
- Source and issues: https://github.com/pyvista/cvista
- PyVista: https://github.com/pyvista/pyvista
9.7.0.5
Sep 10, 2026
9.7.0.4
Sep 05, 2026
9.7.0.3
Aug 25, 2026
9.7.0.2
Aug 23, 2026
9.7.0.1
Aug 23, 2026
9.7.0.0
Aug 22, 2026
9.6.2.6
Aug 16, 2026
9.6.2.5
Aug 14, 2026
9.6.2.4
Aug 12, 2026
9.6.2.3
Jul 06, 2026
Wheel compatibility matrix
| Platform | CPython 3.11 | CPython >=3.12 (abi3) |
|---|---|---|
| macosx_11_0_arm64 | ||
| manylinux_2_28_aarch64 | ||
| manylinux_2_28_x86_64 | ||
| win_amd64 |
Files in release
cvista-9.7.0.5-cp311-cp311-macosx_11_0_arm64.whl (21.5MiB)
cvista-9.7.0.5-cp311-cp311-manylinux_2_28_x86_64.whl (25.7MiB)
cvista-9.7.0.5-cp311-cp311-win_amd64.whl (17.2MiB)
cvista-9.7.0.5-cp312-abi3-macosx_11_0_arm64.whl (21.5MiB)
cvista-9.7.0.5-cp312-abi3-manylinux_2_28_aarch64.whl (21.1MiB)
cvista-9.7.0.5-cp312-abi3-manylinux_2_28_x86_64.whl (22.8MiB)
cvista-9.7.0.5-cp312-abi3-win_amd64.whl (17.2MiB)