Accelerate molecular discovery with physics-based machine learning.

Our technology enables thousandfold speed-ups in molecular prediction with transformative improvements in accuracy.

Learn More
Protein and molecules

Machine learning with quantum featurization

  • Achieve 1000-fold speed up in quantum molecular simulation with OrbNet, without compromising accuracy
  • Use the power of quantum featurization to generate accurate property predictions, even with small datasets
  • Work with us to build predictive models while protecting your IP through anonymizing featurizations
  • Learn More

Qcore: physics-based simulation engine

  • Use Qcore to efficiently perform a full range of electronic structure calculations and simulations
  • Run DFT, fast semi-empirical methods, robust geometry optimization, ab initio dynamics, and properties
  • Integrate Qcore into complex workflows, running locally, on a server, or deployed in the cloud
  • Learn More
Qcore and Envision on a MacBook

Entos for education and academic science

Our technology is applicable to


New ML models can replace DFT in pharma, with 1000x speedups

Read paper


Understanding energy flow in collisions of hydrogen on graphene

Read paper


Multiscale condensed-phase simulation for polyolefin catalysis

Read paper

Let's talk about how we can help your business

We are passionate about science, just like you. We work with clients to provide the best software and machine learning products for their discovery and development needs. Contact us to learn how we can help.

Contact us