DNA LM

DNALM is building next-generation foundation models for genomic data — enabling accurate prediction of phenotypic traits from full DNA sequences across all forms of life. Our core technology, GENA-LM, is a transformer-based DNA language model trained on over 1 trillion base pairs, capable of processing genomic contexts up to 36,000 base pairs. It has demonstrated state-of-the-art performance across more than 20 genomic tasks, including promoter and enhancer identification, variant effect prediction, and gene expression modeling. We offer our platform as an API-accessible SaaS solution for pharmaceutical companies, research labs, and agri-biotech — supporting applications in gene therapy, disease risk assessment, genome annotation, and synthetic biology. Open-source adoption: 80,000+ downloads Active collaborations in bio R&D and wet-lab validation Founded by leading experts in AI, genomics, and molecular biology HQ: United States, with R&D teams in the UK and EU We aim to make genome-scale phenotypic prediction practical, accurate, and scalable. Learn more: dnalm.ai