Download DROPPS

This page provides direct download links for official DROPPS releases. Each package is distributed as a Python .whl file and can be installed directly using pip.

Available Versions

The following versions of DROPPS are currently available:

Installation

You are highly recommended to create a new Python virtual environment before installing DROPPS to avoid conflicts with other packages. You can use tools like venv, conda, or virtualenv to manage your environments.

conda create -n dropps-4.0
conda activate dropps-4.0

To install any version, download the corresponding .whl file and run:

pip install /path/to/dropps-<version>-py3-none-any.whl

Example:

pip install dropps-0.4.0-py3-none-any.whl

Or directly from the URL: .. code-block:: bash

This command will automatically download the wheel file and install it in your current environment. The installation process will also install any required dependencies, such as OpenMM and Numpy, if they are not already present.

GPU support

Normally, the pip install command will automatically install the appropriate OpenMM version for your system, and if your system has a compatible NVIDIA GPU and CUDA drivers, you should have GPU support.

However, if you want to ensure GPU support or if you encounter issues with the default installation (common problem on some high performance computing clusters), you can manually install OpenMM with CUDA support before installing DROPPS:

conda create -n dropps-4.0-gpu
conda activate dropps-4.0-gpu

conda install -c conda-forge openmm=8.1 cudatoolkit
conda install nvidia::cuda-nvcc

pip install https://dropps.online/download/dropps-0.4.0-py3-none-any.whl

Compatibility Notes

  • DROPPS uses openmm as computing backends. Because different versions of openmm may have different APIs, each DROPPS release is built and tested with a specific version of openmm (8.1.*). Please check the release notes for compatibility information before installing.

  • All wheels are built for Python 3.x (py3-none-any).

  • DROPPS is compatible with macOS (intel and Apple Silicon) and Linux. Tests have been performed on CentOS Stream 7/8/9 and macOS 24/25/26.

  • GPU acceleration requires a functional OpenMM-cuda installation.

  • Some analysis scripts use MDAnalysis, NumPy, GSD, or HOOMD-blue.

Latest Release

The current recommended stable release is:

DROPPS 0.4.0

You can install it directly:

pip install https://dropps.online/download/dropps-0.4.0-py3-none-any.whl

Older Releases

Older versions are provided for reproducibility or to match earlier projects. If possible, new users should use the latest version.

Feedback

If you encounter issues during installation, please see:

or open an issue on GitHub.