tech
December 15, 2025
Millions of new materials discovered with deep learning
We share the discovery of 2.2 million new crystals – equivalent to nearly 800 years’ worth of knowledge. We introduce Graph Networks for Materials Exploration (GNoME), our new deep learning tool th…
TL;DR
- The AI tool GNoME has identified 2.2 million new crystal structures, significantly expanding the known materials.
- Of these, 380,000 are predicted to be stable and promising for applications in areas such as superconductors and next-generation batteries.
- GNoME utilizes deep learning and graph neural networks to predict material stability with high accuracy, increasing the discovery rate from around 50% to 80%.
- This AI-driven discovery is equivalent to approximately 800 years' worth of traditional research.
- External researchers have already experimentally created 736 of these predicted structures, validating GNoME's findings.
- The discovered materials' data, particularly the 380,000 stable compounds, will be contributed to the Materials Project database.
- Collaborative efforts are also exploring autonomous synthesis of these new materials using AI-guided robotics.
Continue reading
the original article