NVIDIA and the UW Medicine Institute for Protein Design debuted RFD4-Proteina on October 21, 2026, at GTC Berlin, introducing a new open model that combines programmable protein design with advanced search capabilities to accelerate drug discovery and biological research.
RFD4-Proteina Launches at GTC Berlin
Proteins execute essential molecular tasks throughout biology, but finding a protein capable of a specific designed function requires searching an immense range of possibilities. To address this, NVIDIA and the UW Medicine Institute for Protein Design released RFD4-Proteina as part of the NVIDIA BioNeMo open platform. The model generates targeted protein designs, letting scientists and automated agents search and refine candidates using molecular inputs, constraints, and scoring objectives.
Potential applications include engineering proteins to interrupt disease processes or detect signs of illness in diagnostic tests. Across four benchmark suites—protein binding, small-molecule binding, enzyme scaffolding, and DNA binding—the research draft noted that RFD4-Proteina surpassed evaluated baselines in both computational design success rates and generation throughput during testing.
We are moving from an era of discovering the proteins nature has made to one in which we can design entirely new ones,
stated the University of Washington’s Institute for Protein Design director and 2024 chemistry Nobel Laureate David Baker. RFD4-Proteina expands what scientists can create, from proteins that could advance drug discovery to enzymes that break down pesticides.
Access to the model is provided through various partners, such as Amazon Bio Discovery, Anthropic’s Claude Science, Benchling, Databricks, Google Cloud, Hugging Face, Nebius, OpenAI, and Tamarind Bio.
Siemens Healthineers Advances Radiotherapy With NVIDIA Intelligence
Siemens Healthineers is collaborating with NVIDIA to develop the next generation of adaptive radiotherapy, combining imaging and oncology expertise with accelerated computing and simulation. By 2050, approximately 35 million people worldwide are projected to be diagnosed with cancer each year, increasing pressure on care teams to deliver precise treatments while managing limited specialist capacity.
Rapid patient-specific simulation is being pursued through NVIDIA CUDA-accelerated dose calculation, alongside the NVIDIA Holoscan platform for integrating live sensor data with artificial intelligence, as investigated by the companies.
Adaptive radiotherapy is built on a single principle: treatment should adapt to the patient, not the other way around,
said Arthur Kaindl, head of Varian, the Siemens Healthineers cancer-care business.
Scientific Agents Connect Drug Design and Autonomous Laboratories
Global scientific research and development spending reaches $3.8 trillion, while annual pharmaceutical budgets approach $300 billion. To convert this investment into evidence and new treatments, research technology companies are using the NVIDIA BioNeMo platform and NVIDIA Nemotron models to build scientific agents that connect design, experimentation, and learning.
Benchling is grounding agents in experimental data, while Biolevate is connecting biological data and specialized tools across drug development. Boltz is developing models to predict molecular structures and previewing an agent to plan workflows. Automata is connecting scientific agents with automated laboratory infrastructure to initiate experiments and capture structured results. Danaher’s autonomous lab incorporates RFD4-Proteina within Genedata’s artificial intelligence co-scientist workflow to coordinate protein design and testing. Hologen is combining large medicine models with NVIDIA Nemotron to build clinical scientists, and Therna Biosciences is using large language model reasoning to connect experimental evidence with RNA medicine design.
Hospital Operations Streamline Across European Healthcare Systems
European healthcare systems are working to expand access to specialized care while reducing administrative burdens on clinical staff. Avelios, Corti, Dedalus, Dawn Health, and Sword are building on NVIDIA Nemotron to power applications ranging from automated medical scribes to personalized patient support services. PolyAI is using the NVIDIA Agent Toolkit to accelerate care access via multilingual dialog deployment with medical software provider Corilus, which serves more than 70,000 healthcare professionals in Belgium and France. Apian and VitVio are extending these capabilities into physical hospital operations by using NVIDIA Cosmos world models and NVIDIA Omniverse libraries to build digital twins for monitoring and operational improvement.
Frequently Asked Questions About NVIDIA GTC Berlin Healthcare Announcements
What specific benchmarks did RFD4-Proteina surpass in testing?
According to research draft test results, RFD4-Proteina outperformed evaluated baselines by delivering superior generation throughput and higher computational design success rates across the enzyme scaffolding, small-molecule binding, protein binding, and DNA binding benchmark suites.
Which cloud and AI partners offer access to RFD4-Proteina?
Anthropic’s Claude Science, Amazon Bio Discovery, Benchling, Databricks, Google Cloud, Hugging Face, Nebius, OpenAI, and Tamarind Bio are among the partners through which RFD4-Proteina is obtainable.
How many healthcare professionals are spanned by Corilus in Europe?
Medical software provider Corilus maintains a network spanning more than 70,000 healthcare professionals across Belgium and France.
Worth a look