Insilico Medicine Announces AI Foundation Model Partnership and China Drug Development Deal

Insilico Medicine and Liquid AI announced a partnership that has produced LFM2-2.6B-MMAI (v0.2.1), a lightweight scientific foundation model for pharmaceutical research. The single 2.6 billion-parameter model achieves state-of-the-art performance across multiple drug discovery subdomains while running entirely on private pharmaceutical infrastructure.

The model covers the complete discovery loop, spanning property prediction and ADMET endpoints, multi-parameter molecular optimization, target-aware scoring with protein-pocket conditioning, functional group reasoning, and retrosynthesis planning. Training involved approximately 120 billion tokens of pharmaceutical data across over two hundred different tasks.

At just 2.6B parameters, the model achieved cloud-scale performance while operating entirely on private infrastructure. In property prediction, it outperformed TxGemma-27B, a model more than 10 times larger, on 13 of 22 tasks covering pharmacokinetics and toxicology, and achieved state-of-the-art results on three of these tasks when compared to specialist models built for individual tasks. The model reached success rates of up to 98.8% on industry-standard multi-parameter optimization benchmarks (MuMO-Instruct).

On Insilico's internal affinity prediction benchmark, featuring 2.5 million experimental measurements across 689 protein targets, the model produced better correlation scores than frontier models including GPT-5.1, Claude Opus 4.5, and Grok-4.1. The model also demonstrated strong functional group reasoning capabilities (FGBench) and high-quality single-step retrosynthesis suggestions (ChemCensor metric).

The partnership combines Liquid AI's efficient LFM architecture with Insilico's MMAI Gym, a comprehensive training platform with over 1,000 pharmaceutical benchmarks. MMAI Gym for Science is a domain-specific training environment designed to elevate general-purpose and frontier Large Language Models into pharmaceutical-grade engines for drug discovery and development. The Gym utilizes specialized tracks for Chemical Superintelligence (CSI) and Biology/Clinical Superintelligence (BSI) to teach models domain-specific reasoning across medicinal chemistry, biology, and clinical planning.

Separately, Insilico Medicine entered a drug development partnership with China Medical System Holdings, focusing on central nervous system and autoimmune diseases. The commitment includes co-development of at least two programs, with R&D funding of up to several tens of millions of HK$ per program.

The company selected ISM5059 as a new AI-discovered preclinical candidate, adding another asset to its pipeline. Insilico also received a $5 million milestone payment after oncology compound MEN2501 moved into Phase 1 studies.

Insilico Medicine is a clinical-stage biotechnology company using AI for drug development across cancer, fibrosis, immunity, central nervous system diseases, and aging-related conditions. The company's AI platform covers target discovery, molecular design, and clinical development. Liquid AI builds Liquid Foundation Models based on dynamical systems and signal processing, focusing on AI models that are efficient and can be deployed on-premise or in resource-constrained environments.