SuperConGenAI

What is SuperConGenAI?

SuperConGenAI is an advanced AI-driven model that generates novel superconducting materials based on target critical temperature (Tc). Trained on the SuperCon dataset, it predicts candidate compositions that meet specified Tc values, enabling data-driven exploration of high-Tc superconductors. Designed to accelerate materials discovery, it supports research and development in condensed matter physics, quantum materials, and superconducting technologies.

Key Features

• Generate superconductors based on target critical temperature (Tc)
• Predict critical temperature (Tc) from material composition
• Built-in pre-trained AI models
• Offline support
• No Python installation required

How to Install the App

SuperConGenAI is available as a standalone desktop application for both macOS (Apple Silicon) and Windows. You can download the installers above.

SuperConGenAI for macOS (Apple Silicon)

1. Download the SuperConGenAI for macOS.
2. Open the downloaded file, then follow the setup instructions in README.txt for Mac.
3. After installation, launch SuperConGenAI from Applications.
4. Go to the Superconductors Generator or read the Documentation to get started.

Requirements

• macOS 13 Ventura or later
• Apple Silicon Architecture (M1, M2, M3, M4 or later)
• 8 GB RAM minimum (16+ GB recommended)

Dataset

The SuperCon dataset served as the primary training dataset. SuperCon provides a large collection of experimentally measured superconductors with reported critical temperatures (Tc) by the Materials Database Group, National Institute for Materials Science, 2021.

Disclaimer

This project is intended for educational and research exploration purposes only. The generated superconductors are produced by AI models trained on experimental data from the SuperCon.

While these models capture statistical trends, the resulting compositions are not guaranteed to correspond to experimentally known, synthesizable, or thermodynamically stable superconductors. All outputs should be treated as hypothetical candidates, subject to further validation through computational or experimental methods.

Contact

Feel free to reach out for scientific collaborations or questions about this work.

References

1. National Institute for Materials Science. MDR SuperCon: Superconducting Material Database. 2021. https://doi.org/10.48505/nims.3837