SAN FRANCISCO, Jan. 12, 2026 — Leads & Copy — NVIDIA has announced a significant expansion of its NVIDIA BioNeMo™ platform, designed to foster AI-driven advancements in biology and drug discovery through lab-in-the-loop workflows.
The BioNeMo platform now includes new NVIDIA Clara™ open models like RNAPro for RNA structure prediction and ReaSyn v2 to ensure the feasibility of synthesizing AI-designed drugs. It also incorporates BioNeMo Recipes for scalable biological foundation model training and data processing libraries like nvMolKit, a GPU-accelerated cheminformatics tool for molecular design.
The life sciences sector is known for producing vast quantities of scientific data, and BioNeMo is engineered to handle this data to accelerate discovery while reducing R&D costs, currently estimated at $300 billion annually. The platform aims to transform data into a competitive advantage, maximizing the likelihood of successful outcomes.
Kimberly Powell, vice president of healthcare at NVIDIA, emphasized the transformative potential of AI in biology and drug discovery. She stated that BioNeMo converts experimental data into actionable intelligence, creating a continuous learning cycle that accelerates discovery and supports the development of new models to address complex biological challenges.
NVIDIA is collaborating with leading life sciences organizations to integrate BioNeMo into laboratory experiments and scientific workflows. This collaboration aims to create a full AI lifecycle for biology and drug discovery, effectively linking experimentation with AI.
Lilly and NVIDIA have announced a collaboration to launch a co-innovation lab focused on addressing drug discovery challenges. Thermo Fisher also announced a collaboration with NVIDIA to develop intelligent scientific instruments and autonomous laboratories.
The collaboration between NVIDIA and Lilly will combine NVIDIA’s AI, accelerated computing, and robotics expertise with Lilly’s drug discovery and development capabilities. This integration will support Lilly’s chemists and biologists, leveraging the NVIDIA BioNeMo platform and Lilly’s agentic lab. The companies will also explore the application of accelerated computing and advanced AI across Lilly’s operations, from manufacturing to commercial activities.
According to Diogo Rau, executive vice president and chief information and digital officer at Lilly, this collaboration aims to catalyze the next era of drug discovery by uniting extensive compute capabilities, specialized talent, and data processing at scale. Rau envisions a future where discovery is driven by rapid experimentation and customized models, reflecting Lilly’s commitment to applied AI in drug discovery.
Thermo Fisher’s collaboration with NVIDIA seeks to transform scientific research labs into automated data factories by integrating NVIDIA’s AI computing with Thermo Fisher’s instrumentation. This includes unified edge-to-cloud AI compute using the NVIDIA DGX Spark™ desktop supercomputer, multi-agent systems for lab orchestration using the NVIDIA NeMo™ software suite, and autonomous data analysis with BioNeMo tools.
Gianluca Pettitti, executive vice president of Thermo Fisher Scientific, noted that combining AI with laboratory automation will revolutionize scientific work. The collaboration aims to help customers accelerate discoveries with greater accuracy and efficiency.
The BioNeMo platform is supported by a global ecosystem of innovators who are developing AI for drug discovery. Model builders such as Basecamp Research, Boltz PBC, Chai Discovery, and Natera are using BioNeMo to scale their model training and development.
Companies like Edison Scientific, Tetrascience, and Owkin are developing domain-specific agents for science using NVIDIA open models and the NVIDIA NeMo framework. Additionally, companies like Multiply Labs, Lila Sciences, HighRes Biosolutions, and Opentrons Labworks are integrating simulation and physical AI technologies using NVIDIA Isaac Sim™.
Janette Ciborowski
NVIDIA Corporation
+1-734-330-8817
jciborowski@nvidia.com
Source: NVIDIA
