Search results

Faster D3 for MACE

Status: Open.

#1340 In ACEsuit/mace;

· LesTuche opened on Jan 26, 2026

Reactivity in condensed phase: putting higher weights on solutes' atoms compared to solvents' atoms in the training loss.

Status: Open.

#1339 In ACEsuit/mace;

· ines-art opened on Jan 25, 2026

Memory problem when trying to train with large system.

Status: Open.

#1337 In ACEsuit/mace;

· jungsdao opened on Jan 23, 2026

[Bug] OpenEquivariance (OEQ) integration fails with TypeError in MACE 0.3.14

Status: Open.

#1336 In ACEsuit/mace;

· evenfarther opened on Jan 23, 2026

config_weight functionality is missing in loss functions

Status: Open.

#1335 In ACEsuit/mace;

· vigneshpai2003 opened on Jan 22, 2026

Problems about multihead replay finetuning.

Status: Open.

#1333 In ACEsuit/mace;

· SaltyLemonWHU opened on Jan 21, 2026

using mace_active_learninng_md faces matrix multiplication for finetuned model

Status: Open.

#1325 In ACEsuit/mace;

· Asif-Iqbal-Bhatti opened on Jan 7, 2026

CI improvements & speedups

Status: Open.

#1319 In ACEsuit/mace;

· stenczelt opened on Dec 30, 2025

Compilation Issues of LAMMPS and MACE on CentOS 7 with GPU

Status: Open.

#1318 In ACEsuit/mace;

· shiswang opened on Dec 29, 2025

LAMMPS pair_style mace uses neighbor/ghost cutoff larger than r_max

Status: Open.

#1315 In ACEsuit/mace;

· Pourasad opened on Dec 28, 2025

mace_run_train distributed processes overwrite each others' pretrained data files

Status: Open.

#1309 In ACEsuit/mace;

· bernstei opened on Dec 18, 2025

MACE-LES with MLIAP not working with LAMMPS

Status: Open.

#1307 In ACEsuit/mace;

· nhleocn opened on Dec 15, 2025