Metadata-Version: 2.4
Name: ovstage
Version: 0.1.1.355824
Summary: NVIDIA OVStage Data Plane Python API
Author: NVIDIA Corporation
License-Expression: LicenseRef-NvidiaProprietary
Project-URL: Homepage, https://github.com/NVIDIA-Omniverse/ovstage
Project-URL: Documentation, https://nvidia-omniverse.github.io/ovstage/
Project-URL: Repository, https://github.com/NVIDIA-Omniverse/ovstage
Project-URL: Changelog, https://github.com/NVIDIA-Omniverse/ovstage/blob/main/CHANGELOG.md
Keywords: nvidia,omniverse,ovstage,usd,openusd
Classifier: Development Status :: 3 - Alpha
Classifier: Intended Audience :: Developers
Classifier: Intended Audience :: Science/Research
Classifier: Natural Language :: English
Classifier: Programming Language :: C
Classifier: Programming Language :: C++
Classifier: Programming Language :: Python :: 3.10
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Classifier: Programming Language :: Python :: 3.13
Classifier: Programming Language :: Python :: 3 :: Only
Classifier: Environment :: GPU :: NVIDIA CUDA
Classifier: Operating System :: OS Independent
Classifier: Topic :: Scientific/Engineering
Classifier: Topic :: Software Development :: Libraries
Requires-Python: >=3.10
Description-Content-Type: text/markdown
License-File: LICENSE
Dynamic: license-file

# NVIDIA ovstage

[![Documentation](https://img.shields.io/badge/docs-ovstage-blue)](https://nvidia-omniverse.github.io/ovstage)
[![Repository](https://img.shields.io/badge/github-NVIDIA--Omniverse%2Fovstage-181717)](https://github.com/NVIDIA-Omniverse/ovstage)

**ovstage** is a C and Python library providing a shared, high-performance, vectorized, GPU-capable scene data substrate for [OpenUSD](https://openusd.org) scene data, for use in Omniverse Libraries spanning physics, rendering, sensors, animation, and more.

It provides C and Python APIs for reading, writing, querying, and managing simulation data such as transforms, velocities, materials, hierarchy, and metadata across CPU and GPU memory, with zero-copy data paths and DLPack tensor interchange. Currently, zero-copy applies to CUDA source-tensor writes; payload reads and map/unmap buffers are CPU-resident. No Omniverse background is needed — it is a standalone library that gives your application a fast, vectorized, in-memory representation of a scene to read and write from C or Python.

> **Pre-release:** the API, runtime behavior, and packaging may change.

## Install

```
pip install ovstage
```

## Documentation

Full C and Python API documentation, guides, and runnable examples are at
**<https://nvidia-omniverse.github.io/ovstage>**. Source, examples, and the C SDK
release archives live at <https://github.com/NVIDIA-Omniverse/ovstage>.

## License

NVIDIA Proprietary Software. See the LICENSE file included in this distribution.
