---
id: vesin
version: "0.6.1"
license: BSD-3-Clause
license_treatment: permissive
maintenance: active
---

# vesin — Computing neighbor lists for atomistic system

License: permissive · Maintenance: active · Downloads: 351.7K/mo

## What it is and what it does

Vesin is a compiled library for computing neighbor lists in atomistic simulations—the core operation of identifying which atoms interact within a specified cutoff distance. It wraps a high-performance C implementation and exposes it via a Python API, with support for periodic boundary conditions and various output formats (pair indices, shift vectors, distances). The library is designed for molecular dynamics, machine learning on atomic structures, and other computational chemistry workflows where neighbor identification is a bottleneck.

You can use it either through its own NeighborList class or via a drop-in replacement for ASE's neighbor list function, making it suitable for integration into existing atomistic simulation pipelines. It requires numpy at runtime and Python 3.10 or later.

Use it for:

- Computing pairwise interactions in molecular dynamics simulations with periodic boundary conditions.
- Generating training data for machine learning models on atomic structures by identifying neighbor pairs.
- Accelerating force-field calculations in computational chemistry by quickly finding atoms within cutoff range.
- Integrating into ASE-based workflows as a faster drop-in replacement for neighbor list computation.
- Building graph representations of atomic systems for neural network potentials.

## Worth the install?

AI-flagged interpretation of the facts above — verify before relying.

Vesin computes neighbor lists for atomistic systems—identifying which atoms are within a cutoff distance of each other—with a Python interface backed by compiled code for speed.

Yes, if you work with atomistic simulations or machine learning on atomic structures. Vesin solves a specific, performance-critical problem with active maintenance, no known vulnerabilities, and permissive licensing. The medium install friction is offset by precompiled wheels for common platforms and a straightforward API. Not relevant for non-atomistic applications.

## Install

pip install vesin
uv add vesin
poetry add vesin

## Installing vesin

Before you install: Medium install friction due to precompiled wheels for common platforms (macOS arm64/x86_64, Linux x86_64/aarch64, Windows x86_64), but requires numpy at runtime. Repository is active with recent commits and a stable release cadence.

License in practice: BSD-3-Clause is permissive; you can use, modify, and distribute vesin freely in commercial or private projects provided you include the license notice and disclaim liability.

Quickstart:

pip install vesin

import numpy as np
from vesin import NeighborList

positions = [(0, 0, 0), (0, 1.3, 1.3)]
box = 3.2 * np.eye(3)
calculator = NeighborList(cutoff=4.2, full_list=True)
i, j, S, d = calculator.compute(points=positions, box=box, periodic=True, quantities="ijSd")

Requires Python >=3.10 and numpy as a runtime dependency.

Verify before relying:

- Whether GPU acceleration (mentioned for NVIDIA H100 in benchmarks) is available in the Python interface or only in C/C++.
- Performance characteristics compared to ASE's native neighbor list for typical system sizes.

## Package facts

- License: BSD-3-Clause (permissive)
- Python support: supports_current
- Install friction: medium
- Maintenance: active
- Downloads: 351.7K/month (top 15,000 on PyPI)
- Known vulnerabilities: none known

## Tags

neighbor list computation, atomistic neighbor search, molecular dynamics neighbor pairs, cutoff-based atom pairing, periodic boundary neighbor list, fast neighbor detection atoms, molecular-dynamics, computational-chemistry, neighbor-search

[View on SkillFed](https://skillfed.io/packages/vesin) · [View on PyPI](https://pypi.org/project/vesin/)