mace

Kaldi model converter to ONNX

A curated list of awesome inference deployment framework of artificial intelligence (AI) models. OpenVINO, TensorRT, MediaPipe, TensorFlow Lite, TensorFlow Serving, ONNX Runtime, LibTorch, NCNN, TNN, MNN, TVM, MACE, Paddle Lite, MegEngine Lite, OpenPPL, Bolt, ExecuTorch.

DeePMD-kit plugin for various graph neural network models

Interface for simulations and standalone Python script generation with universal machine-learning interatomic potentials (MACE, CHGNet, SevenNet, NequIP, ORB, Allegro, MatterSim, UPET, GRACE, UMA, ALIGNN-FF)

High-performance CUDA kernels for equivariant graph neural networks (MACE, NequIP, Allegro). 10-20x faster than e3nn.

easy build for XiaoMI/mace

🔧 Python package for 3rd/4th-order force constant calculation via finite displacement method, with ML potential support & thermal disorder generation.

SimpleAtom, a web interface for MACE(Message Passing Atomic Cluster Expansion) a machine-learning-based interatomic potential

A Python package for amorphous structure generation by melt-quench simulation and random placement.

pose estimation for android using mace engine

MACE is A C++ Engine

A fast style transfer application based on the Xiaomi's MACE framework (基于小米MACE框架的快速风格迁移应用)

This respository contains workflows to calculate transition state other relevant properties using QuAcc library.

MLIP (Machine Learning Interatomic Potential) plugins for ORCA ExtTool (ProgExt) interface.

A Lightweight high-performance symmetric encryption framework designed to provide Full Diffusion

Run machine-learning potentials using VASP style inputs.

Python module for adsorption structure generation and potential energy surface scans of molecule–surface systems using Bayesian Optimization (BOSS)-guided structure generation and MACE machine-learned potentials.

MLIP (Machine Learning Interatomic Potential) plugins for Gaussian 16 External interface.

Transfer learning using a high order message passing foundational model to F-Li-B cells.

Autonomous active-learning pipeline (MACE + jobflow) that trains an MLIP to DFT-level accuracy, automates CI-NEB calculations and extracts Brønsted–Evans–Polanyi scaling relations for H₂ dissociation on N-doped graphene-supported Pd single-atom catalysts.