vesin-torch
Computing neighbor lists for atomistic system, in TorchScript
Decision gist · record as of 2026-08-14
Yes, if you work with atomistic simulations or machine learning on molecular systems and use torch. The package is actively maintained, has no security vulnerabilities, carries a permissive license, and solves a real performance bottleneck. Medium install friction is acceptable given the platform-specific wheels and single torch dependency. Not necessary if you already have adequate neighbor list performance or don't use torch.AI-flagged interpretation of the facts on this page — verify before relying
Before you install
- Requires Python >=3.10 and torch as a runtime dependency.
- Medium install friction due to platform-specific wheels and a single runtime dependency on torch.
- Actively maintained with recent releases; last commit 2026-08-07 and version 0.6.1 released 2026-07-29.
License · maintenance · safety
BSD-3-Clause (permissive) — Distributed under BSD-3-Clause (permissive), allowing commercial and private use with attribution and liability disclaimers. No restrictive copyleft obligations.
last release 2026-07-29 (16 days) · last repo commit 2026-08-07 · 85 stars
0 known vulnerabilities (OSV.dev, 2026-08-14) · 263,606 downloads/mo, #8,353 on PyPI
Alternatives
Verify before relying
pip install vesin-torch
from vesin import NeighborList
positions = [(0, 0, 0), (0, 1.3, 1.3)]
calculator = NeighborList(cutoff=4.2, full_list=True)
i, j, S, d = calculator.compute(points=positions, periodic=True, quantities="ijSd")- Whether numpy is required as a transitive dependency or only for the usage example.
- GPU performance characteristics and hardware requirements beyond torch support.
- Compatibility with specific torch versions or CUDA/ROCm backends.
- Whether TorchScript interface is accessible from the Python package or requires separate build.
What it is and what it does
Vesin-torch is a library for computing neighbor lists in atomistic simulations—a fundamental operation in molecular dynamics and materials science where you need to identify which atoms are within a cutoff distance of each other. It wraps the Vesin C library and exposes it through PyTorch, enabling both CPU and GPU execution. The package provides a straightforward Python API via the NeighborList class, which takes atomic positions and a cutoff distance, then returns pairs of neighbor indices and their separation vectors, optionally accounting for periodic boundary conditions.
The library is designed for performance on large systems. It also offers drop-in compatibility with existing workflows. The single runtime dependency is torch, and the package is actively maintained with recent releases and no known security vulnerabilities.
Use it for
- Accelerating molecular dynamics simulations by computing neighbor lists on GPU with torch.
- Building machine learning models for atomistic systems that need efficient neighbor pair enumeration.
- Computing pairwise distances and indices for large crystal supercells with periodic boundaries.
- Prototyping interatomic potential models that require fast neighbor list updates.
- Integrating into existing chemistry workflows that need neighbor list performance improvements.
Worth the install?
AI-flagged interpretation of the facts on this page. Verify before relying on it.
Yes, if you work with atomistic simulations or machine learning on molecular systems and use torch.
The package is actively maintained, has no security vulnerabilities, carries a permissive license, and solves a real performance bottleneck. Medium install friction is acceptable given the platform-specific wheels and single torch dependency. Not necessary if you already have adequate neighbor list performance or don't use torch.
Install
vesin-torch on PyPI
Before you install
Medium install friction due to platform-specific wheels and a single runtime dependency on torch. Actively maintained with recent releases; last commit 2026-08-07 and version 0.6.1 released 2026-07-29.
Requires Python >=3.10 and torch as a runtime dependency.
License in practice
Distributed under BSD-3-Clause (permissive), allowing commercial and private use with attribution and liability disclaimers. No restrictive copyleft obligations.
Quickstart
pip install vesin-torch
from vesin import NeighborList
positions = [(0, 0, 0), (0, 1.3, 1.3)]
calculator = NeighborList(cutoff=4.2, full_list=True)
i, j, S, d = calculator.compute(points=positions, periodic=True, quantities="ijSd")Verify before relying
- Whether numpy is required as a transitive dependency or only for the usage example.
- GPU performance characteristics and hardware requirements beyond torch support.
- Compatibility with specific torch versions or CUDA/ROCm backends.
- Whether TorchScript interface is accessible from the Python package or requires separate build.
Package facts
| License | BSD-3-Clause permissive |
| Python support | Supports the current Python release >=3.10 |
| Install friction | Medium. Platform-specific wheel |
| Runtime dependencies | 1 packagetorch |
| Maintenance | Actively maintained 16 days since the last release |
| Last repo commit | |
| First released | |
| Downloads | 263,606 / month, #8,353 on PyPI 30-day window, as of 2026-08-14 |
| Known vulnerabilities | None known OSV.dev, checked 2026-08-14 |
| Classifiers | Development Status :: 4 - BetaIntended Audience :: Science/ResearchOperating System :: MacOS :: MacOS XOperating System :: Microsoft :: WindowsOperating System :: POSIXProgramming Language :: PythonProgramming Language :: Python :: 3Topic :: Scientific/EngineeringTopic :: Scientific/Engineering :: Bio-InformaticsTopic :: Scientific/Engineering :: ChemistryTopic :: Scientific/Engineering :: PhysicsTopic :: Software Development :: LibrariesTopic :: Software Development :: Libraries :: Python Modules |
Evidence: vesin_torch-0.6.1-py3-none-macosx_11_0_arm64.whl; vesin_torch-0.6.1-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl; vesin_torch-0.6.1-py3-none-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl; vesin_torch-0.6.1-py3-none-win_amd64.whl
Tags
Let your AI agent find packages like this
Example. Real query, live index.
You found this page by searching. An agent finds it by wishing: SkillFed indexes 14,416 PyPI packages by what they can do, searchable in plain language.
wish › “atomistic systems”
- vesin-torchComputes neighbor lists for atomistic systems efficiently, with…
- vesinVesin computes neighbor lists for atomistic systems—identifying which…
- nvalchemi-toolkit-opsGPU-accelerated batched primitives for atomistic simulation: neighbor…
Give your agent the search over MCP, or paste the wish link into any chat.
More Libraries packages
urllib3 is an HTTP client library that provides thread-safe connection pooling, SSL/TLS verification, multipart file uploads, request retries, compression support, and proxy handling for Python applications.
Requests is a Python HTTP library that simplifies sending HTTP/1.1 requests with automatic handling of headers, authentication, cookies, and response parsing.
Pluggy provides a plugin system that lets you define hook specifications and register implementations to be called in sequence, enabling extensible Python applications without tight coupling.
Install it if you're building an extensible application or framework.
Provides parsing, arithmetic, and recurrence rule computation for dates and times, with timezone support and iCalendar RFC compliance.
Install it if you need to parse flexible date strings, compute relative dates, handle timezones, or work with recurrence rules—it's the de facto choice for these tasks.
Six provides utility functions to write Python code that runs on both Python 2.7 and Python 3.3+, smoothing over language differences between the two versions.
pytest is a testing framework that lets you write test functions using plain assert statements and automatically discovers and runs them, with detailed failure reporting.
See also nvalchemi-toolkit-ops · vesin · matscipy · chgnet · cudensitymat-cu13 · ase · fairchem-core · cuvs-cu12 · libcuvs-cu12 · dscribe