Vote in Free Arena

News··1 min read·Data updated automatically

NVIDIA releases cuPotential library to speed up molecular dynamics on GPUs

cuPotential offers modular C API primitives that double to triple MD simulation speed on GPUs from T4 to RTX PRO 6000 Blackwell.

The library provides drop‑in replacements for OpenMM’s nonbonded force, Langevin integrator and Monte‑Carlo barostat, requiring minimal code changes.

Benchmarks show end‑to‑end speedups of 2× to over 3× on GPU architectures ranging from T4 to RTX PRO 6000 Blackwell for systems of 24,000 to 400,000 atoms under NPT conditions.

New continuous splitting and spreading functions, co‑developed with D. E.

Energy drift over 10 ns simulations is reduced by several hundredfold versus default PME settings, while dispersion energy accuracy improves by several thousandfold with comparable force errors.

What it means for you

Researchers can obtain faster MD runs without sacrificing accuracy, enabling larger or longer simulations on existing GPU hardware. Developers of free or locally hosted AI tools can integrate cuPotential via its OpenMM plugin to boost performance in scientific workloads that rely on molecular dynamics.

Sources

Written automatically from the sources above, then checked by our code: every number and name must appear in them, and no sentence may copy a source. Spotted a mistake? Report it.

Free leaderboard

Which AI can you use for free today?

Every free AI model, ranked by blind human votes, with the memory and speed it needs on your own GPU.

Open the leaderboard →

Share this page:XLinkedInRedditWhatsApp

Press / or Ctrl+K to search from any page.