1. [CUDA](/cuda) [CUDA-X Libraries](/cuda/cuda-x-libraries/) NVIDIA ALCHEMI # NVIDIA ALCHEMI: AI Lab for Chemical and Materials Innovation NVIDIA ALCHEMI is a collection of domain-specific [NVIDIA NIM](https://build.nvidia.com)™ microservices and a toolkit for accelerating chemical and materials discovery for diverse use cases, such as battery materials, catalysts, organic light-emitting diodes (OLEDs), and beauty products. Discovery in these areas is historically slow and costly due to the trial-and-error nature of experimentation. Traditional computational chemistry methods are either not accurate enough or too computationally expensive. By accelerating AI-enabled atomistic simulation at the machine learning framework level, ALCHEMI enables researchers and developers to achieve performant simulations at near-quantum chemistry accuracy, speeding up chemical and materials discovery. [![Click to Watch on Youtube](https://developer.download.nvidia.com/images/products/chemistry-materials-science.jpg "Click to Play")](https://youtu.be/O-2TLhaTcuw?si=i8r18EvMrpwqUlWR%20) * * * ## How ALCHEMI Works NVIDIA ALCHEMI delivers capabilities across three integrated layers: - **ALCHEMI NIM microservices:** A scalable layer of cloud‑ready, domain‑specific microservices for chemistry and materials science, enabling deployment and orchestration on NVIDIA‑accelerated platforms - **ALCHEMI Toolkit:** A collection of GPU-accelerated simulation building blocks, including geometry optimizers, integrators, and data structures, to enable large-scale, batched simulations with AI - **ALCHEMI Toolkit-Ops:** A repository of GPU-accelerated, batched common operations for AI-enabled atomistic simulation tasks, such as neighbor list construction, DFT-D3 dispersion corrections, and long-range electrostatics ALCHEMI NIM​ is intended for researchers who need atomistic simulation at near-quantum chemistry accuracy to inform chemical and materials discovery​. ALCHEMI Toolkit​ and Toolkit-Ops are intended for community code, independent software vendor (ISV), and startup developers training machine learning interatomic potentials (MLIPs) and building accelerated batched MLIP atomistic simulation engines​. ![Diagram showing an overview of NVIDIA ALCHEMI.](https://developer.download.nvidia.com/images/products/accelerate-simulation-graph.svg) * * * ## Get Started With ALCHEMI ### ALCHEMI NIM Scalable, cloud‑ready microservices for chemistry and materials science [Download from NGC Catalog](https://catalog.ngc.nvidia.com/)[Batched Geometry Relaxation](https://catalog.ngc.nvidia.com/orgs/nim/teams/nvidia/containers/alchemi-bgr?version=1.0.0)[Batched Molecular Dynamics](https://catalog.ngc.nvidia.com/orgs/nim/teams/nvidia/containers/alchemi-bmd?version=1.0.0) ### ALCHEMI Toolkit Modular training, simulation, and active learning workflows [Download on GitHub](https://github.com/NVIDIA/nvalchemi-toolkit)[Read Documentation](https://nvidia.github.io/nvalchemi-toolkit) **Installation With PyPI** pip install nvalchemi-toolkit **Installation With uv** uv venv --seed --python 3.12 && source .venv/bin/activate uv pip install nvalchemi-toolkit ### ALCHEMI Toolkit-Ops GPU-accelerated, batched kernels for AI-enabled atomistic simulation [Download on GitHub](https://github.com/NVIDIA/nvalchemi-toolkit-ops)[Read Documentation](https://nvidia.github.io/nvalchemi-toolkit-ops/) **Installation With PyPI** pip install nvalchemi-toolkit-ops **Installation With Conda** conda create -n nvalchemi python=3.12 pip conda activate nvalchemi pip install nvalchemi-toolkit-ops **Installation With uv (Recommended for Maintainers)** uv venv --seed --python 3.12 uv pip install nvalchemi-toolkit-ops * * * ## Starter Kits ### ALCHEMI NIM Access cloud‑ready, domain‑specific microservices for chemistry and materials science to deploy and orchestrate on NVIDIA‑accelerated platforms. - [Read Quick Start Guide: Batched Geometry Relaxation (BGR)](https://docs.nvidia.com/nim/alchemi/alchemi-bgr/latest/quickstart-guide.html) - [Read Quick Start Guide: Batched Molecular Dynamics (BMD)](https://docs.nvidia.com/nim/alchemi/alchemi-bmd/latest/quickstart-guide.html) - [Read Documentation: Batched Geometry Relaxation (BGR)](https://docs.nvidia.com/nim/alchemi/alchemi-bgr/latest/index.html) - [Read Documentation: Batched Molecular Dynamics (BMD)](https://docs.nvidia.com/nim/alchemi/alchemi-bmd/latest/index.html) ### ALCHEMI Toolkit Create your own custom simulation workflows with GPU-accelerated building blocks, including geometry optimizers, integrators, and data structures. - [Read Quick-Start Guide](https://nvidia.github.io/nvalchemi-toolkit/userguide/about/install.html) - [Read ALCHEMI Toolkit Documentation](https://nvidia.github.io/nvalchemi-toolkit/modules/index.html) ### ALCHEMI Toolkit-Ops Leverage GPU-accelerated, batched common operations for AI-enabled atomistic simulation tasks, such as neighbor list construction, DFT-D3 dispersion corrections, and long-range electrostatics. - [Read Quick-Start Guide](https://nvidia.github.io/nvalchemi-toolkit-ops/userguide/about/install.html) - [Read ALCHEMI Toolkit-Ops Documentation](https://nvidia.github.io/nvalchemi-toolkit-ops/modules/index.html) * * * ## ALCHEMI Learning Library Blog NVIDIA Accelerated Computing Enables Scientific Breakthroughs for Materials Discovery NVIDIA ALCHEMI boosts research for OLEDs and cooling fluids by enabling high-throughput, AI-driven screening of billions of candidate molecules. Blog NVIDIA ALCHEMI NIM Catalyzes Sustainable Materials Research ALCHEMI NIM microservices optimize AI inference for chemical simulations, significantly speeding the discovery of sustainable energy materials for EV batteries, solar panels, and more. Tech Blog Faster Chemistry and Materials Discovery With AI-Powered Simulations Using NVIDIA ALCHEMI Batched Conformer Search and Molecular Dynamics NIM microservices rapidly identify stable molecule shapes and simulate complex material behaviors. Tech Blog Accelerating AI-Powered Chemistry and Materials Science Simulations With NVIDIA ALCHEMI Toolkit-Ops ALCHEMI Toolkit-Ops provides GPU-accelerated PyTorch kernels for essential atomistic operations like neighbor list construction and long-range electrostatics. Tech Blog Revolutionizing AI-Driven Material Discovery Using NVIDIA ALCHEMI ALCHEMI accelerates chemical discovery using AI, featuring batched geometry relaxation to speed up material stability simulations by 100x. Video How Accelerated Compute Is Driving Scientific Discovery From OLEDs to battery electrolytes, ALCHEMI accelerates chemical and materials discovery using CUDA-X™. * * * ## More Resources ![Decorative image representing forums](https://developer.download.nvidia.com/icons/m48-developer-1.svg) ### Join the NVIDIA Developer Program ![](https://developer.download.nvidia.com/icons/m48-email-settings.svg) ### Sign Up for the Developer Newsletter