
MatterSim is a deep learning atomistic model that works across elements, temperatures and pressures. It is distributed as a Python package with pre-trained models.
Web appFree
Step 1: Create a clean conda environment with a supported Python version.
Step 2: Install the package with pip install mattersim.
Step 3: Build an atomic structure with ASE and attach the MatterSim calculator.
Step 4: Read the potential energy, forces and stress from the structure.
Step 5: Optionally switch to the larger checkpoint or run the fine-tuning script on your own data.
The smaller, faster one. You can load the larger, more accurate one by setting the load path on the calculator.
Yes. The package includes a fine-tuning script that trains a pre-trained model on a custom dataset using torchrun.
Yes, but on Apple Silicon the README recommends the CPU device because the MPS backend can show numerical instability.
Cite the MatterSim preprint and state the exact model version and checkpoint you used, not just the name MatterSim.
MatterGenMatterGen is a generative model from Microsoft for designing inorganic materials across the periodic table. It can be fine-tuned to steer generation toward property constraints.
BiomedCLIPBiomedCLIP is a biomedical vision-language model from Microsoft, pretrained on 15 million figure-caption pairs from PubMed Central. It supports retrieval, image classification and visual question answering research.
AuroraAurora uses AI for AI foundation model for Earth-system prediction across weather, air quality, ocean, and environmental forecasting tasks.
CHGNetCHGNet is a pretrained universal neural network potential for charge-informed atomistic modeling. It predicts energy, force, stress and magnetic moments of crystal structures at near DFT accuracy.
OmegaFoldProtein structure prediction model using a protein language model without requiring multiple sequence alignments
Humanloop EvalsHumanloop Evals is an AI-native product for evaluation and prompt-management platform for testing and improving LLM applications with datasets, evaluators, an.