OSS Tanbou
Sign in with GitHub
View on GitHubProject website

Embed rigid-body and collision simulation in games and visualization tools

OSS scale score 204.4OSS health 93Update pace 1 commits/30d
About these scores

OSS scale score is an unbounded metric that log-compresses and weights Stars, Watchers, Forks, and Contributors. Discovery score is the current OSS scale score minus the score at discovery. Update pace is commits in the last 30 days, growth momentum is the OSS scale score difference within the recent observation window, and OSS health is a 0–100 rating based on available recency, Community Health, and release data.

Stars
4,784
Primary language
C++
License
BSD-3-Clause
Repository last updated
Sep 18, 2026
On this page

Overview

This repository contains three lines: ovphysx, the PhysX SDK, and Omniverse PhysX Extensions. This article targets the native PhysX SDK 5.11.0 under physx/, a C++ SDK for rigid bodies, collision detection, joints, vehicles, and related simulation. Its core features run on CPUs, while selected features can use supported NVIDIA GPUs.

Features and best fit

Based on official documentation; not hands-on tested · Content checked:

Key features

Compute contacts and motion through one simulation model

Applications combine shapes, materials, rigid bodies, and joints to compute contacts and constrained motion over time. Included snippets and examples provide small starting points for individual SDK features.

Sources: [2][3]

Choose CPU builds or optional GPU features

The public presets include a Linux configuration that disables GPU support and builds the core SDK and snippets for CPU use. With a supported CUDA environment, applications can also select GPU-accelerated rigid-body features.

Sources: [2][4]

Best fit

Use it where applications need controlled physical behavior

PhysX fits games, training simulators, robotics visualization, and digital-twin applications that need to construct and control a physical scene. It is an SDK rather than a complete engine, so rendering, input, and asset management come from the host application.

Sources: [1][2]

Before adoption

Check the supported OS, compiler, and C++ requirements

Repository tag ovphysx-0.6.3 contains native PhysX SDK 5.11.0 under physx/. Its Linux instructions require x86-64, glibc 2.31 or later, CMake 3.21 or later, Python 3.5 or later, curl, and C++14, and list compiler combinations for Ubuntu 22.04 and 24.04. Project generation uses Packman to download dependencies.

Sources: [3]

Review the root and SDK license scopes separately

GitHub reports BSD-3-Clause from the repository root, while the SDK under the physx directory carries Apache License 2.0. Before distribution, review notices for the exact directories and bundled components used by the product.

Sources: [5][6]

Official sources

  1. [1]NVIDIA-Omniverse/PhysX README at ovphysx-0.6.3(2026-10-04)
  2. [2]PhysX SDK README at ovphysx-0.6.3(2026-10-04)
  3. [3]PhysX Linux build instructions at ovphysx-0.6.3(2026-10-04)
  4. [4]PhysX Linux Clang CPU-only preset at ovphysx-0.6.3(2026-10-04)
  5. [5]NVIDIA-Omniverse/PhysX root license at ovphysx-0.6.3(2026-10-04)
  6. [6]PhysX SDK license at ovphysx-0.6.3(2026-10-04)
  7. [7]PhysX ovphysx-0.6.3 release(2026-10-04)
Supplemental curator note

Evaluate the API and toolchain first with a small CPU-only snippet, then decide which GPU features and rendering stack the application needs.

Try it in 3 steps

  1. 1

    Fetch the pinned SDK release

    Clone repository release ovphysx-0.6.3 and enter physx/, which contains native PhysX SDK 5.11.0. The ovphysx release number is not the SDK version.

    git clone --branch ovphysx-0.6.3 --depth 1 https://github.com/NVIDIA-Omniverse/PhysX.git physx-demo && cd physx-demo/physx
  2. 2

    Generate a CPU build configuration

    On supported Linux with Clang, CMake, Python, and curl, generate the public preset with GPU features disabled. Packman downloads build dependencies.

    ./generate_projects.sh linux-clang-cpu-only
  3. 3

    Build the HelloWorld snippet

    From the same physx directory, build the minimal CPU snippet without rendering, a GPU, or external hardware.

    cmake --build compiler/linux-clang-cpu-only-release --target SnippetHelloWorld --parallel 2
Check the official README

Growth

Growth trends · Last 30 days

4,784 Stars

Trend data is still being collected.

Development activity

Last 90 days · weekly

Commits (last 30 days)
1
Open PRs
26
Issues opened
40
Issues closed
13
PRs opened
25
PRs merged
0

Issues

40 / 13

Jul 8Oct 5
Issues openedIssues closed

Pull requests

25 / 0

Jul 8Oct 5
PRs openedPRs merged

Maintenance

Median first response
24 days
Issue response rate
2.5% (1/40)

Based on up to the 100 newest issues opened by external users in the last 90 days. A first comment from an OWNER, MEMBER, or COLLABORATOR counts as a response; issues whose full comment history cannot be checked are excluded. The median and response rate update weekly.

Built with

Categories and tags

Categories

Tags

GitHub data

GitHub dataView detailed GitHub data
View on GitHub
Stars
4,784
Forks
671
Watchers
68
Open issues
91
Contributors
11
Owner type
Organization
Primary language
C++
Repository last updated
Sep 18, 2026
Write a related article

Share a guide or use case for this OSS in Markdown. Articles are published after administrator approval.

Report incorrect information

Tell us if any listing information is incorrect or outdated.

After reading this page, do you know what to do next?