The ibm/biomed.omics.bl.sm.ma-ted-458m model family is a biomedical foundation model and its finetuned variants trained on over 2 billion biological samples across multiple modalities, including proteins, small molecules, and single-cell gene data. Designed for robust performance, it achieves state-of-the-art results over a variety of tasks across the entire drug discovery pipeline and the diverse biomedical domains.

Based on the MAMMAL - Molecular Aligned Multi-Modal Architecture and Language, a flexible, multi-domain architecture with an adaptable task prompt syntax. The syntax allows for dynamic combinations of tokens and scalars, enabling classification, regression, and generation tasks either within a single domain or with cross-domain entities.

This page demonstrates a variety of drug discovery and biomedical tasks for the model family. Select the task to access the specific demos.