{"id":27463,"library":"torch-sim-atomistic","title":"torch-sim-atomistic","description":"A PyTorch toolkit for calculating material properties using machine learning interatomic potentials (MLIPs). Current version 0.6.0 supports SIMPL, MACE, CHGNet, and other models via a unified interface. Release cadence is quarterly.","status":"active","version":"0.6.0","language":"python","source_language":"en","source_url":"https://github.com/shoshansw/torch-sim-atomistic","tags":["machine-learning","interatomic-potentials","materials-science","pytorch","MLIP"],"install":[{"cmd":"pip install torch-sim-atomistic","lang":"bash","label":"pip"}],"dependencies":[{"reason":"Core tensor operations and autograd","package":"torch","optional":false},{"reason":"Atomic simulation environment for structure handling","package":"ase","optional":false}],"imports":[{"note":"Old package name was torch-sim; now torch-sim-atomistic","wrong":"from torch_sim import TorchSim","symbol":"TorchSim","correct":"from torch_sim_atomistic import TorchSim"},{"note":"Models submodule under torch_sim_atomistic.models","symbol":"MACE","correct":"from torch_sim_atomistic.models import MACE"}],"quickstart":{"code":"import torch\nfrom torch_sim_atomistic import TorchSim\nfrom ase.build import bulk\n\n# Create a silicon crystal\natoms = bulk('Si', 'diamond', a=5.431)\n\n# Initialize MLIP calculator\ncalc = TorchSim(model='MACE', device='cpu')\natoms.calc = calc\n\n# Compute energy\nenergy = atoms.get_potential_energy()\nprint(f\"Energy: {energy:.3f} eV\")","lang":"python","description":"Compute potential energy of bulk silicon using a MACE model."},"warnings":[{"fix":"Update imports: from torch_sim_atomistic import ... instead of from torch_sim import ...","message":"In v0.6.0, the package name changed from 'torch-sim' to 'torch-sim-atomistic'. Import paths updated accordingly.","severity":"breaking","affected_versions":"<0.6.0 -> 0.6.0+"},{"fix":"Switch to supported models: TorchSim(model='MACE') or TorchSim(model='SIMPL')","message":"Models like CHGNet and M3GNet are deprecated in v0.6.0 and will be removed in v0.7.0. Use MACE or SIMPL instead.","severity":"deprecated","affected_versions":"0.6.0"},{"fix":"Always set device explicitly: TorchSim(..., device='cpu') or device='cuda'.","message":"The default device is 'cuda' if available; this can cause issues on CPU-only systems without explicit device setting.","severity":"gotcha","affected_versions":"all"}],"env_vars":null,"last_verified":"2026-04-27T00:00:00.000Z","next_check":"2026-07-26T00:00:00.000Z","problems":[{"fix":"Install the correct package: pip install torch-sim-atomistic, and import from torch_sim_atomistic.","cause":"Package was renamed from torch-sim to torch-sim-atomistic in v0.6.0.","error":"ModuleNotFoundError: No module named 'torch_sim'"},{"fix":"Install with MACE extra: pip install 'torch-sim-atomistic[mace]' or install MACE separately: pip install mace-torch.","cause":"MACE model might not be installed or registered. Requires optional dependency: pip install torch-sim-atomistic[mace].","error":"ImportError: cannot import name 'MACE' from 'torch_sim_atomistic'"},{"fix":"Use supported models: 'MACE' or 'SIMPL'. For CHGNet, downgrade to v0.5.x.","cause":"CHGNet and M3GNet were deprecated in v0.6.0 and removed.","error":"ValueError: Unknown model 'CHGNet'. Available models: ['MACE', 'SIMPL']"}],"ecosystem":"pypi","meta_description":null,"install_score":null,"install_tag":null,"quickstart_score":null,"quickstart_tag":null}