Overview
CREST, the Conformer-Rotamer Ensemble Sampling Tool, automates exploration of low-energy molecular chemical space. The README describes it as an OpenMP scheduler for efficient force-field and semiempirical methods such as xTB, with additional capabilities for creating and analysing structure ensembles.
Installation options include statically linked binaries from the release page, conda-forge packages, and CMake or meson builds from source. Since version 3.0 the primary run types use the integrated tblite library, so xTB is no longer strictly required for them, while features such as QCG still need an xTB installation. Ensembles therefore reflect the sampling method and the underlying force-field or semiempirical level, not a DFT result.
COPYING and COPYING.LESSER state LGPL-3.0. The project documentation site carries installation instructions and application examples.
Key Features
- Automated conformer and rotamer ensemble sampling with ensemble creation and analysis options.
- Acts as an OpenMP scheduler for force-field and semiempirical methods such as xTB, with tblite integrated since version 3.0.
Use Cases
- Build a ranked conformer ensemble before thermochemistry or property calculations.
How to Use
Unpack a release binary and add it to the path, install the conda-forge package, or build with CMake or meson as described in the documentation. Check whether the chosen run type needs an xTB installation; the README notes that only some functions still require it.
See the official documentation for installation and examples: https://crest-lab.github.io/crest-docs/
Sources and verification
This entry describes capabilities documented by the linked project sources. Catalog edits are not evidence of a successful run, compatibility test or scientific performance evaluation.
Read the editorial and verification policy