NVIDIA has launched new AI-powered tools within its ALCHEMI (AI Lab for Chemistry and Materials Innovation) suite to accelerate chemistry and materials discovery. Announced on November 18, 2025, the NVIDIA Batched Conformer Search (BCS) NIM and NVIDIA Batched Molecular Dynamics (BMD) NIM microservices are designed to optimize atomistic simulations, crucial for predicting chemical properties and stability. The ALCHEMI BCS NIM efficiently identifies and ranks low-energy conformers of molecules, using AIMNet2 as a machine learning interatomic potential (MLIP) to accelerate energy optimization. This reduces the time required to generate low-energy conformers compared to traditional quantum chemistry methods. The BMD NIM facilitates high-throughput molecular dynamics simulations through dynamic batching and GPU-based integrators, supporting various MLIPs like MACE-MPA-0 and TensorNet, which enables concurrent processing and maximizes throughput. Key features of the BMD NIM include:

Dynamic batching: Optimizes GPU utilization by dynamically batching atomic systems.
GPU-based integrators: Performs simulations at a constant number of atoms, volume, and temperature (NVT), or a constant number of atoms, pressure, and temperature (NPT).
MLIP support: Supports MACE-MPA-0, TensorNet-MatPES-r2SCAN-v2025.1, TensorNet-MatPES-PBE-v2025.1, AIMNet2, AIMNet2-NSE, and AIMNet2-CPCM.

Universal Display Corporation (UDC) is leveraging ALCHEMI BCS and BMD NIM microservices to accelerate OLED molecular discovery, predicting thermal processing stability up to 10,000x faster than traditional density functional theory (DFT). ENEOS has also utilized the NIMs, along with the batched DFT microservice, to discover over 1,000 promising environmentally friendly and non-PFAS immersion cooling fluids in just three weeks. If you're at Supercomputing 25, more information can be found at the NVIDIA booth, and you can check out the NVIDIA ALCHEMI demo.