Skip to content

Chai-1

Predict molecular structures for drug discovery

Visit official site

What Chai-1 is

The model is described as a multimodal foundation model for drug discovery, predicting structures of proteins, small molecules, DNA, RNA and covalent modifications. The material reports a 77% PoseBusters success rate.

What you can do with it

  • Predict a molecular structure from sequences
  • Choose a single-sequence or MSA mode
  • Add lab data as a prediction prompt

What to watch out for

  • Web access follows registration and login
  • A prediction mode is selected before running
  • Weights and code are for non-commercial use

Pros & cons

✓ What we like

  • Lab data prompts can improve prediction
  • The web interface is listed as free, including commercial use

! What to watch out for

  • Registration and login are needed for web access
  • Weights and inference code are released for non-commercial use

FAQ

What benchmark figure is listed?

The material states 77% on PoseBusters.

Is there a web interface?

A free web interface at lab.chaidiscovery.com is listed.

Last reviewed: 2026-11-15

More LLM API platform tools

View all →

How we review