# Chai-1 / Chai-2

Source: https://onco.cc/technologies/chai-1/  
OnCo record `chai-1` (Technology). Data CC BY-NC 4.0, attribute "Data from OnCo (onco.cc)"; commercial use needs a licence.

## TL;DR

Structure and antibody-design models from Chai Discovery, with Chai-2 reporting high zero-shot antibody hit rates.

## Summary

Chai-1 and Chai-2 are structure prediction and protein design models from Chai Discovery. Chai-1 (2024) is a diffusion model for biomolecular structure prediction released with open weights, and Chai-2 (2025) adds generative design conditioned on a chosen target epitope, producing antibodies and other binders from scratch. Chai-2 reported roughly 16% zero-shot binder hit rates across dozens of targets in wet-lab tests, meaning a meaningful fraction of computer-designed antibodies bound their target without experimental optimisation. The intended users are biologics teams seeking new antibodies, including against cancer antigens. Independent replication of the Chai-2 hit rates is pending, and developability of the designed antibodies has not been reported. For a newcomer: Chai-2 claims it can design a working antibody from a target's structure on the first try in about one attempt in six.

## Fields

- Kind: Technology
- Status: emerging
- Last checked: 2026-09-08
- Tags: foundation-model; protein-design
- Principle: Diffusion structure prediction; generative design conditioned on target epitope.
- Since: 2024
- Strengths: De novo antibody design
- Limitations: Independent replication pending

## Sources

- Chai Discovery: https://www.chaidiscovery.com

## Connected records

- fronts: [AI & Computation](https://onco.cc/fronts/ai-computation/), [Drug Discovery Platforms](https://onco.cc/fronts/drug-discovery/)
- technologies: [AI-driven drug & target discovery](https://onco.cc/technologies/ai-drug-design/), [Monoclonal antibodies](https://onco.cc/technologies/monoclonal-antibody/)
- companies: [Chai Discovery](https://onco.cc/companies/chai-discovery/)
- roadmaps: [AI in oncology roadmap: pattern readers → foundation models → agents in the workflow](https://onco.cc/roadmaps/ai-oncology-roadmap/), [Drug discovery roadmap: screening in mice → maps of dependency → designing in silico](https://onco.cc/roadmaps/drug-discovery-roadmap/)

---
JSON: https://onco.cc/api/v1/entities/chai-1.json