{"version":1,"type":"story","url":"https://digestai.news/story/nvidia-and-university-of-washington-unveil-open-source-protein-design","json":"https://digestai.news/story/nvidia-and-university-of-washington-unveil-open-source-protein-design.json","markdown":"https://digestai.news/story/nvidia-and-university-of-washington-unveil-open-source-protein-design.md","slug":"nvidia-and-university-of-washington-unveil-open-source-protein-design","headline":"Nvidia and University of Washington unveil open-source protein design model RFD4-Proteina","summary":"Nvidia and the University of Washington School of Medicine Institute for Protein Design have released RFD4-Proteina, an open‑source generative model built on Nvidia’s BioNeMo platform. The developers say the model improves design success rates by five‑ to ten‑fold and generates candidates five‑ to ten‑times faster than earlier tools. Benchmarks across protein binding, small‑molecule binding, enzyme scaffolding and DNA binding suites showed higher computational success and throughput, according to Nvidia.\n\nRFD4-Proteina is distributed under a BSD‑3 license and hosted on a BSD‑3 repository. It can be accessed through partners such as Amazon Bio Discovery, Anthropic’s Claude Science, Benchling, Databricks, Google Cloud, Hugging Face, Nebius, OpenAI and Tamarind Bio, enabling integration into cloud and scientific workflows. Funding for the project came from the Howard Hughes Medical Institute, the Gates Foundation (grant of $25,748,619 dated November 2025), Grantham Foundation, Fund for Science and Technology, Schmidt Sciences, the OpenFold consortium and Rosetta Commons. Nobel laureate David Baker highlighted the model’s potential to accelerate drug discovery, enzyme engineering and broader biotechnology applications.","keyPoints":["RFD4-Proteina achieves 5‑10× higher design success rates than prior models, developers say.","Generation speed is 5‑10 times faster, cutting design time from months to days.","The model is open‑source under a BSD‑3 license and available via multiple cloud partners."],"whyItMatters":"An open, high‑performance protein‑design AI lowers barriers for drug and enzyme discovery, letting researchers iterate designs days instead of months and fostering broader collaboration across biotech and sustainability fields.","category":{"slug":"models","name":"Generative AI & Models","url":"https://digestai.news/category/models"},"entities":{"companies":["Nvidia","University of Washington","Amazon Bio Discovery","Anthropic","Google Cloud","OpenAI"],"models":["RFD4-Proteina","BioNeMo","RosettaFold Diffusion","Proteina-Complexa","La-Proteina","Claude Science"],"people":["David Baker","Meg Lunn-Halbert","Seth Woodbury","Nathaniel Corley","Maxwell Kazman","Rafi Brent"]},"firstPublishedAt":"2026-10-22T10:48:00Z","updatedAt":"2026-10-22T10:48:00Z","sourceCount":1,"hasPrimarySource":false,"sources":[{"outlet":"genengnews.com","title":"Nvidia, UW Institute for Protein Design Unveil Open-Source Protein Design AI Model","url":"https://genengnews.com/topics/artificial-intelligence/nvidia-uw-institute-for-protein-design-unveil-open-source-protein-design-ai-model","publishedAt":"2026-10-22T10:48:00Z","type":"press","primary":false,"lead":true}],"sourceNotes":null,"discussions":[],"thread":null,"cite":{"text":"Digest AI, \"Nvidia and University of Washington unveil open-source protein design model RFD4-Proteina\", 22 October 2026, https://digestai.news/story/nvidia-and-university-of-washington-unveil-open-source-protein-design","publisher":"Digest AI","title":"Nvidia and University of Washington unveil open-source protein design model RFD4-Proteina","datePublished":"2026-10-22T10:48:00Z","url":"https://digestai.news/story/nvidia-and-university-of-washington-unveil-open-source-protein-design"},"generatedBy":"Written by Digest AI's editorial model from the linked sources; the sources are the record.","license":"Headlines, digests and key points are written by Digest AI and may be quoted with a link to the story page. Linked articles belong to their publishers. Terms: https://digestai.news/terms#reuse"}