Model Medicines announced that founder and CEO Daniel Haders II, PhD, will present at the Artificial Intelligence Convergence: Small Molecule Discovery Summit, held March 19-20, 2026, in Boston, Massachusetts. He is scheduled to lead a roundtable, Building Agents & Exploring Emerging LLM Use Cases for Small Molecule Discovery Funnel Applications, on Thursday, Mar 19, 2026, and deliver a presentation, Breaking the Throughput Barrier: Ultra-Large Virtual Screening as a Precision Amplifier for Artificial Intelligence-Driven Drug Discovery, on Friday, Mar 20, 2026.
The company argues that inference throughput is the main bottleneck limiting the reach of Artificial Intelligence-driven drug discovery into deeper chemical space. Laboratory-based High-Throughput Screening (HTS) has the potential to evaluate one million compounds per day. State-of-the-Art (SOTA) Artificial Intelligence-Driven drug discovery campaigns are capable of screening two million (BioHive) to eight billion (Atomwise) compounds per day. Together, laboratory-based and current SOTA Artificial Intelligence-driven screenings have been constrained to a ceiling of 8E+9 in chemical space, while all potential drug-like compounds are estimated to exist in a chemical space of 1E+60. Model Medicines said this gap contributes to incremental advances, rediscovery of known scaffolds, and limited access to novel chemistry in unexplored regions.
Model Medicines said its Ultra-Large Virtual Screening architecture is designed to overcome that constraint. The company executed a 325-billion-compound ULVS in a day in 2025 in partnership with Google. It described that effort as the largest machine-learning-driven bioactivity screen publicly reported to date. Touching 3E+11 chemical space, this approach was the foundation for the development of two first-in-category programs against the “undruggable” transcription factor BRD4 and the novel broad-spectrum RdRp Thumb-1 target. The company also said it is constructing a one-trillion-compound (1E+12) scale screen.
The roundtable focus will shift from screening scale to multi-parameter optimization, which the company describes as drug discovery’s defining challenge. Model Medicines said every program is guided by a Target Product Profile that links molecule design to patient needs and development constraints, spanning efficacy, safety, dosage, administration, storage, market access, regulatory milestones, and manufacturing feasibility. It said translating that profile into a molecule requires optimizing affinity, potency, selectivity, solubility, oral bioavailability, tissue partition, half-life, pharmacokinetics, drug-drug interactions, ADME properties, synthesizability, and stability at the same time.
As an example of that approach, the company pointed to AmesNet, which it recently published and released as an agent that replaces the regulatory required Ames genotoxicity test. Model Medicines said AmesNet outperformed all literature-reported Ames agents, including the FDA’s DeepAmes, Baidu’s GROVER, and MIT’s ChemProp. The company said agent-based systems and emerging LLM applications could be used across the discovery funnel to accelerate decisions, coordinate across disciplines, and keep programs aligned with the Target Product Profile. Beyond platform work, Model Medicines said its pipeline includes MDL-001, a direct-acting, non-nucleoside, broad-spectrum antiviral, and MDL-4102, a BRD4 inhibitor with no measurable activity against BRD2/3.
