Periodic Labs introduced Periodic Neon, a 1-trillion parameter model trained on proprietary laboratory data to outperform existing models on scientific analysis tasks like X-ray diffraction.

A deep dive into the company's philosophy on building autonomous labs and AI scientists to accelerate material discovery and restart Moore's Law.

Periodic Labs officially emerged from stealth with a $300 million seed round and a mission to build AI scientists that automate physical laboratory experiments.

An in-depth interview with co-founder Liam Fedus exploring the company's technical approach to training AI on private experimental traces rather than scientific literature.