MatCraft - AI-Powered Materials Discovery

Accelerating materials science research with surrogate models, active learning, and multi-objective optimization.

🎯 What we publish here

• Datasets: 205k+ unified materials database (Materials Project + AFLOW + JARVIS)
• Models: Pre-trained surrogate models for 16 material domains
• Demos: Interactive Gradio spaces for materials discovery

🔗 Links

• Website: https://matcraft.ai
• Documentation: https://matcraft.ai/docs

📊 Use Cases

• ML researchers: Clean datasets with 30+ normalized properties
• Materials scientists: Fast property predictions without DFT
• Students & educators: Free access to real materials data

📝 Citation

If you use MatCraft data or models in your research, please cite:

@misc{matcraft2026, title={MatCraft: AI-Powered Materials Discovery Platform}, author={MatCraft Team}, year={2026}, url={https://matcraft.ai} }

📧 Contact

Questions? → contact@matcraft.ai