METiS TechBio, a trailblazer in AI-powered nano-delivery innovation, officially launched NanoForge—the world’s first comprehensive AI-driven nano-delivery platform. The unveiling was led by Dr. Chris Lai, Co-Founder and CEO of METiS, who, along with a multidisciplinary team of experts in AI, platform engineering, and RNA research, showcased NanoForge’s groundbreaking capabilities and its real-world applications in advancing therapeutic pipeline development.

NanoForge represents the first fully integrated AI platform for nano-delivery discovery and design. Leveraging quantum chemistry and molecular dynamics simulations, it combines METiS' proprietary high-throughput wet-lab screening platform with in-house developed generative lipid language models, AI agents for nanomaterial research, and the world's largest LNP lipid library—containing over 10 million unique lipid structures. The platform supports a closed-loop workflow encompassing molecular generation, property prediction, AI-guided dry and wet lab experimentation, formulation design and optimization, and final candidate selection. Through continuous learning, NanoForge steadily enhances its data infrastructure and predictive accuracy. Built on this platform are three core solutions: AiLNP (AI Nucleic Acid Delivery Design), AiRNA (AI mRNA Sequence Design), and AiTEM (AI Small Molecule Formulation Design).

To date, METiS has built a library of over 10 million lipid structures and accumulated 100,000 high-quality data points for model training. The company has achieved breakthrough targeted LNP delivery to eight organs and tissues in vivo, including the liver, lung, immune organs, heart, muscle, tumors, central nervous system, and gastrointestinal tract. METiS holds more than 100 granted and pending patents. It has advanced over 10 pipeline programs, including 7 preclinical candidates, and 4 clinical-stage programs, with the most advanced nearing pre-NDA stage.

NanoForge incorporates four key proprietary technological breakthroughs:

The World's Largest Ionizable Lipid Library
The METiS lipid library—the core of NanoForge's data ecosystem—contains over 10 million lipid structures, with a significant subset validated through wet-lab experiments measuring more than 20 key attributes. This extensive and proprietary dataset supports robust model training, helps circumvent existing IP barriers, and systematically captures both positive and negative results to refine organ-specific targeting.

Specialized Foundational AI Models for Nanomaterials
NanoForge is built on a suite of vertical AI foundation models that span all critical stages of nanomaterial development: molecular generation, property prediction, experimental design, formulation optimization, and validation. These models operate in parallel across functional domains, enabling efficient small-molecule nanomaterial optimization and accurate prediction of up to 20 key physicochemical and functional properties of lipids and lipid nanoparticles.

The Industry's First Nano-Delivery AI Agent
METiS developed ALAN, an AI agent that serves as an intelligent interface for NanoForge. It processes natural language and structured queries to guide users through lipid-based delivery system discovery and optimization. ALAN interprets tasks, decomposes them into sub-tasks, and dynamically engages relevant models and datasets—including the curated lipid library, internal screening data, patents, and external literature. It also incorporates a reasoning layer that integrates multi-model outputs, applies domain-specific rules, and delivers results with transparent and scientifically traceable logic.

Integrated Dry-Wet Lab Infrastructure
NanoForge's AI-guided dry and wet lab system forms a closed loop from in silico design to in vivo validation, enabling rapid iteration and refinement of AI predictions through biological experimentation. This design-learn-verify cycle significantly shortens development timelines and improves clinical translation success. For lipid optimization, NanoForge employs mechanistic simulations, predictive analytics, and in vivo tests to assess candidates based on target delivery and biodistribution profiles, enabling early selection of lipids with optimal therapeutic indices. For formulation screening, METiS' proprietary high-throughput system, METiS DATALOTS, tests hundreds of formulation combinations in parallel, dramatically accelerating the discovery of novel LNPs tailored to specific organs and routes of administration.

Dr. Chris Lai, Co-Founder and CEO of METiS, commented:
"NanoForge embodies a new paradigm—from biotech to techbio—that we have built over the past five years. In the specialized field of nano-delivery, we have proven that combining large-scale AI innovation with closed-loop experimental iteration drives remarkable gains in efficiency and technical capability. Rooted in independent R&D and world-class science, METiS is poised to become a new engine for innovative drug development."