Iambic Launches Enchant v3 – Molecular Superintelligence Designed to Advance End-to-End Drug Discovery & Development
Enchant v3 is the next generation of Iambic's multimodal transformer model, trained using 41 billion parameters on more than 6,000 molecular properties across 16 biomedical data modalities
Enchant v3 ingests diverse datasets, deploys a mixture-of-experts architecture, and tests multiple hypotheses in parallel to optimize discovery and development decisions
SAN DIEGO, September 21, 2026--Iambic, a clinical-stage life science and technology company developing novel medicines using its AI-driven platform, today unveiled Enchant v3, the next generation of its multimodal transformer model built to advance end-to-end drug discovery and development. Enchant is the 'brain' behind Iambic's molecular superintelligence platform that, when combined with Iambic's in-house high-throughput experimental chemistry and biology capabilities, can learn from and predict many properties for a target compound profile. Read the report about Enchant v3 here.
"Enchant v3 is an architectural leap over Enchant v2, with new capabilities to be deployed for internal and partner drug discovery and development efforts," said Fred Manby, PhD, Co-Founder and CTO, Iambic. "Earlier versions of Enchant have demonstrated utility across our pipeline, helping us predict critical preclinical and clinical endpoints. Enchant has shown robust scaling laws, with larger models exhibiting more powerful predictive capabilities. Enchant v3 continues this trend and constitutes our next step to make better technology for better medicines."
Enchant was built to:
Ingest diverse data, including public and proprietary data and different modalities, such as assays, text, images, multi-omics, biomolecular structures and sequences, clinical trial data, and more
Leverage model scale so that Enchant gets better at predicting one endpoint when trained on data from other, related endpoints and on other molecules
Break through data walls that have historically separated preclinical drug discovery from clinical development
Convert model outputs into probability-guided decisions by using uncertainty quantification to maximize useful information from laboratory experiments and prioritize potential compounds
Highlights of Enchant's next generation, v3, include:
41 billion parameters for greater reasoning complexity
Pretrained on 4.5 trillion tokens
6,000+ molecular properties for multi-parameter optimization
16 biomedical modalities, including text, assay data, molecule structure, protein structure, 3D structure, images, multi-omics, biologics, clinical trial data, and pharmacokinetics, for a more comprehensive understanding of the underlying chemistry and biology as well as clinical tractability
Predictable scaling laws, demonstrating that increased model scale correlates with improved prediction accuracy across a diverse range of preclinical and clinical endpoints