Makes Chai-1 and Chai-2, open-weight structure models used for antibody and binder design.
Chai Discovery, based in San Francisco, makes Chai-1 and Chai-2, open-weight structure prediction models used for antibody and binder design. Chai-1, released in 2024, matched AlphaFold 3 on many benchmarks with open weights for non-commercial use, and Chai-2, released in 2025, reported zero-shot de novo antibody hit rates far above earlier methods. OnCo links it to the Chai-1 and Chai-2 technology record. Whether high computational hit rates translate into antibodies that behave in patients is the open question. Chai-1 and Chai-2 have their own technology page.
Shares the tags ai, protein-design.
Shares the tags ai, protein-design.
Shares the tags ai, protein-design.
Shares the tags ai, protein-design.
Shares the tags ai, protein-design.
Shares Drug discovery roadmap: screening in mice → maps of dependency → designing in silico and the tag ai.
Shares Drug discovery roadmap: screening in mice → maps of dependency → designing in silico and the tag ai.
Shares Drug discovery roadmap: screening in mice → maps of dependency → designing in silico and the tag ai.
Open-source projects that this organisation maintains, from OnCo's own catalogue: licence and last activity as the repository reported them on the day of the fetch. Listing is not endorsement; check the licence before reuse and the validation before clinical use.
Chai Discovery's multi-modal structure prediction model; code and weights are released, with commercial use permitted under its terms.