Bioptimus

Bioptimus is an AI company building foundation models of human biology. Its models are trained natively across modalities and scales, from cell to tissue to organ, to learn the dynamics of biology and predict what happens next. The product line is organized around H Optimus and M Optimus models, along with STELA, a multi institution initiative that generates deeply profiled, clinically linked, multimodal patient data to serve as a training atlas. The company sells to pharmaceutical and research organizations. Stated applications include drug target and biomarker discovery, where models predict the spatial distribution of antibody drug conjugate targets, immune microenvironment and stromal architecture from routine histology slides without new assays. Others are indication expansion, which scores archived trial slides to find patient subsets likely to respond, spatial biology and diagnostics that combine histology, spatial transcriptomics and genomics, and treatment response and trial design. The site highlights a team of machine learning researchers and biologists drawn from groups such as Google DeepMind, Meta and Owkin. Documentation, a blog, case studies and publications are linked from the site. Pricing is not published. Engagement appears to be through direct contact with the company rather than self serve plans, and the website has no pricing page.