GROMACS 2025.3 Patch Release

GROMACS 2025.3 Patch Release

GROMACS 2020.3 was released in August 2025. This 3rd patch release fixes several issues found since the previous release of the 2025 series. GROMACS recommends all users of the 2025 series to update to 2025.3.

Here are some highlights, along with more details in the links below!

https://manual.gromacs.org/2025.3/index.html

Fixes where mdrun could behave incorrectly

AWH for free-energy lambda coordinate was not active

When using AWH for a free-energy lambda coordinate (and no other dimension), the lambda state would remain at the initial state and all resulting free energies would be zero. Affected versions: 2024.4, 2024.5, 2025.0, 2025.1 and 2025.2.

AWH bias sharing caused error without separate PME ranks

Using AWH in conjuction with bias sharing between several simulations caused mdrun to segfault if more than one rank per simulation was used but no separate PME ranks were present. This could not produce incorrect results. Affected versions: 2024.5, 2025.0, 2025.1 and 2025.2.

gmx mdrun now throws with input inconsistent with the modular simulator

By default, gmx mdrun can choose the modular simulator if possible and fall back on the legacy simulator if needed. The behaviour can be specified with the use of documented environment variables. After this fix, the use of GMX_USE_MODULAR_SIMULATOR environment variable is properly honored. If the simulation cannot use the modular simulator, an error is given.

Fixes that affect portability

SYCL build now works with BUILD_SHARED_LIBS=off

Previously gmx and some test binaries failed to link to libgromacs.a because the required information to link to the SYCL runtime library is missing in that case. Linking of executables in SYCL builds is now more robust.

Fix build with ROCm 7 and newer

Building and linking GROMACS would fail due to changes in the linking behaviour of the main HIP library.

Building GROMACS would fail due to __AMDGCN_WAVEFRONT_SIZE macro removal in ROCm 7.

Improved configuration error message with SYCL and VkFFT

Configuring GROMACS with SYCL and VkFFT, targeting Intel devices, would result in an error meessage implying using VkFFT with SYCL wasn’t suported at all. The message is now clarified to include the statement that this lack of support only applies to Intel devices.

Fix detection of unusable MKL GPU offloading library during configuration

Configuration would previously happily proceed even if development headers of MKL GPU offloading library were not found, causing the compilation to fail. This now results in an error.

Fix build for NVIDIA sm_101 / sm_110 devices

Miscellaneous

Documented that PME tuning in multi-replica simulations can give suprising results

PME tuning chooses between nominally equivalent model physics, yet there is no guarantee of consistency between (say) the tunings chosen for different replicas in a multi-replica simulation, so the fine details of comparisons of potential energy can be surprising. Disabling PME tuning for e.g. testing Hamiltonian replica-exchange simulations with PLUMED is advised.

mdrun reported double the number of domain decompositions when using GPUs

When using GPUs, the cycle count table at the end of the log file reported a number of calls for Domain decomp. that was too high by a factor of one plus the number of dimensions that were decomposed. The times and cycle counts were correct.

Support added for compilation with NVIDIA CUDA 13 toolkit

Addition of a compilation option necessary to avoid CUDA build failures when using the NVCC compiler from version 13 of the CUDA toolkit.

Avoid installing 3rd party header files

Fixed a regression where GROMACS installed muParser headers, causing conflicts with the separate muParser package if installed.

libgromacs now only links to MPI libraries when explicitly required

Formerly gmxapi could have erroneously triggered GROMACS to compile and link with an MPI library. This is now fixed.

Declared gmxapi option before finding MPI package

Formerly gmxapi was set up after MPI libraries, which caused a subsequent run of CMake to decide to find MPI libraries to help gmxapi even when libgromacs didn’t need those libraries. Now the CMake cache does not vary between successive runs in this way.

Improve heuristics for choice of NBNXM kernel flavor for AMD gfx94x

Analytical Ewald kernels are used since these perform better in most common use-cases. The exception is vdw-modifier=Force-switch with SYCL backend, where we still use tabulated kernels.