Mat3ra.com adds support for NVIDIA ALCHEMI

San Francisco Bay Area, CA, USA – WEBWIRE – Thursday, November 12, 2026

With ALCHEMI on the Mat3ra.com platform, users can perform thousands of explorations in batched mode in minutes.

Mat3ra.com today announced support for the NVIDIA ALCHEMI Toolkit on the Mat3ra.com platform. Researchers can now run batched atomistic simulations on NVIDIA GPUs directly from a browser notebook, with no software to install and no GPU to buy.

AI models can propose new materials faster than anyone can test them. Every candidate must be relaxed into a stable structure before anyone knows whether it is worth making. With the NVIDIA ALCHEMI Toolkit 0.3.0, hundreds of structures can be simulated at once on a single GPU instead of one at a time, and the Mat3ra.com platform makes that workflow available to its community of over 40,000 scientists.

In the first benchmark on the platform, one NVIDIA H100 NVL GPU relaxed crystal structures 40-60x faster than the conventional one-at-a-time approach: 4,000 structures in about three minutes, whereas the conventional route needs close to three hours. The benchmark used the MACE machine-learned interatomic potential accelerated by NVIDIA cuEquivariance. The whole workflow is one Jupyter notebook: users can swap in their own structures, or a dynamics or reaction-path script, and the rest of the workflow stays the same.

“Exploring different atomic pathways, including structural relaxation, temperature dependence, the behavior of defects and impurities, and other effects arising in real-world materials and manufacturing, requires searching enormous phase spaces,” said Timur Bazhirov, CEO of Mat3ra.com. “With ALCHEMI on the Mat3ra.com platform, users can perform thousands of explorations in batched mode in minutes. The workflow is modular: our users plug in their own structures, compositions, and use cases while building on the optimal setup from the NVIDIA ALCHEMI team.”

“The Mat3ra platform is intuitive to use, and flexible to bring ALCHEMI quickly,” said Geetika Gupta, Senior Director Product Management at NVIDIA. “It puts the full performance of the GPU within easy reach of end users, with no environment to set up. ALCHEMI is under very active development, and on Mat3ra users can try its latest features right after each release. As AI models propose more candidates than ever, this lets researchers test them in bulk and reach validated materials faster.”

Benchmark figures are geometry-relaxation measurements on one configuration, one run per case; throughput varies with operation, structures, model, settings, and hardware.

Availability.

Blog post with full results: https://www.mat3ra.com/news-and-blog-posts/mat3ra-adds-support-for-nvidia-alchemi
Notebooks: https://github.com/mat3ra/api-examples
Platform: https://platform.mat3ra.com

About Mat3ra.com

Mat3ra.com is a cloud-based research and development platform driving innovation through the strategic integration of cloud high-performance computing and artificial intelligence for materials science. The platform empowers a community of over 40,000 scientists worldwide to collaborate and generate the high-quality, ground-truth physical data essential for advancing materials AI across the semiconductor, energy, and defense sectors. Mat3ra.com serves a diverse customer base ranging from Global 2000 companies to leading public R&D institutions.

For more information, visit Mat3ra.com.

Media Contact

Team Mat3ra.com press@mat3ra.com