Loading
NVIDIA BioNeMo is a software stack, not a model: GPU-optimised training recipes and inference services on which protein, DNA and single-cell foundation models such as Evo 2, ESM and Geneformer are trained and served. The Arc Institute, Recursion and pharmaceutical companies use it, at the price of being tied to NVIDIA hardware and tooling.
NVIDIA BioNeMo is a software stack rather than a single model: frameworks, training recipes and inference microservices optimised for GPUs that biology foundation models are trained and served with. It supports protein, DNA and single-cell models, including Evo 2, ESM and Geneformer, and is used by the Arc Institute, Recursion and pharmaceutical companies to train large models efficiently. Its role in oncology is indirect, as the infrastructure beneath the structure, genome and cell models that feed drug discovery. Because it is vendor-specific, it ties users to NVIDIA hardware and tooling, a practical trade-off against portability. For a newcomer: BioNeMo is the plumbing that many biology AI models run on, not a model that makes predictions itself.
BioNeMo provides optimised training and inference libraries on GPUs.
Query for this technology: (TITLE:"NVIDIA BioNeMo" OR ABSTRACT:"NVIDIA BioNeMo") AND (cancer OR tumor OR tumour OR oncology OR carcinoma OR lymphoma OR leukemia OR leukaemia OR myeloma OR sarcoma OR melanoma OR glioma). Results are unfiltered search hits about NVIDIA BioNeMo, not a curated reading list.
Shares NVIDIA, AI in oncology roadmap: pattern readers → foundation models → agents in the workflow.
Shares NVIDIA, AI in oncology roadmap: pattern readers → foundation models → agents in the workflow.
Open-source projects that implement or serve this technology, 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.
NVIDIA's framework for training and running biology foundation models at scale.