LoQI

5-30 minutesCodeNeed >5GB

Low-energy molecular conformer generation with quantum-mechanical accuracy

Readiness

Readiness: Need >5GB
Smoke: not_tested
Artifacts: not_applicable
Production: need_gt5gb_gpu_loqi_model_checkpoint_pending
Storage: 5-20GB
Strategy: dedicated LoQI GPU PyTorch/PyG/RDKit SIF + model/checkpoint overlay
Next: Build/stage LoQI GPU SIF with Python 3.10, CUDA-compatible PyTorch/PyG/RDKit/Lightning, LoQI source, loqi.ckpt/loqi_flow.ckpt, and AIMNet2 model; run scripts/sample_conformers.py on tiny SMILES with n_confs=1 through Slurm.

No obvious large model/database dependency identified; validate upstream docs before install. 2026-07-05 SOP batch recheck: backend inventory found no staged LoQI source/checkpoints/AIMNet2 model/GPU SIF/artifacts/runner. Upstream requires Python 3.10+, CUDA-aligned PyTorch/PyG/RDKit/Lightning plus loqi.ckpt, loqi_flow.ckpt, processed assets, and AIMNet2 model; GPU is recommended. Reclassified from Smoke Pending to Need >5GB.