Radical Numerics, an AI research lab building general biological intelligence, today launched Omnii-1, a biological AI model that learns across DNA, RNA, proteins and other molecular data. Researchers used Omnii to design a bacteriophage and administered it to a patient with a drug-resistant bacterial infection, marking the first AI-designed phage given to a human patient.

Omnii-1 is a genome language model built on the same AI technology behind large language models, but trained on biological data instead of human language. Radical Numerics developed a way to convert 28 types of biological data, including DNA, RNA, proteins and epigenetic signals, into a common representation that AI can learn from. At launch, the company is demonstrating Omnii across eight applications, including personalized cancer vaccine design, identifying the genetic causes of rare diseases and detecting emerging biological threats.

From AI-generated DNA to a human patient

Radical Numerics was founded by members of the team behind Evo and Evo 2, models used to design the first bacteriophage genomes created entirely by AI. The research, first reported in August 2026, helped establish that AI could go beyond reading the genetic code to writing new biological systems.

Drug-resistant infections were linked to an estimated 4.95 million deaths worldwide in 2019, including 1.27 million directly attributable to antimicrobial resistance. For patients whose infections no longer respond to antibiotics, few options remain. One alternative is bacteriophages, viruses that kill bacteria. But finding a phage that works against a particular patient’s infection can take months and often fails. Omnii offers a different approach: instead of searching for the right virus, design one.

In a recent case, researchers used Omnii to design phages targeting the patient’s persistent drug-resistant infection within weeks. A cocktail containing an AI-designed phage killed the patient’s bacterial isolate in laboratory testing and was then administered at a clinical site under a physician compassionate use access.

“For most of medicine’s history, we’ve searched for drugs that might work against a disease. We’re entering a future where we can design medicines for the exact biology of a patient,” said Eric Nguyen, co-founder and CEO of Radical Numerics. “But we didn’t build Omnii for any single application. We designed it to be a general-purpose AI model that can learn to tackle virtually any task in biology. The phage reaching a patient is one of the first examples of what that approach can make possible.”

“The hardest part of phage therapy is the blind search, not the biology. You’d screen thousands of candidates and mostly learn which ones don’t work,” said Dr. Marisa Azad, an infectious diseases physician-scientist at The Ottawa Hospital. “Omnii changes what we’re searching with: instead of guessing and testing, we can point it at a resistant strain and get back designs built to match it. That’s the difference between hoping you find a phage and actually designing one, with safety measures built in. For patients with life-threatening, antibiotic-resistant infections, we need to get these therapies to the patients who need them most—and get them there rapidly. That’s where Omnii can make a real difference.”

Applying Omnii to biodefense

Because the same advances that make it possible to design medicines can also be used to engineer pathogens that existing detection methods may miss, Radical Numerics is developing Omnii for biodefense. In testing, the model identified AI-generated pathogen sequences that conventional screening tools missed. Omnii analyzes DNA sequencing data as it arrives and returns results within minutes, enabling near real-time pathogen detection. With a laptop and portable DNA sequencer, this technology could eventually bring biological threat detection directly to battlefields, emergency rooms and other locations where rapid response matters. Radical Numerics is also collaborating with Lawrence Livermore National Laboratory, a national lab playing a vital role in the national security S&T innovation ecosystem.

Advisors guiding the development of Omnii’s biodefense capabilities include Andrew C. Weber, former assistant secretary of defense for nuclear, chemical and biological defense programs; retired Maj. Gen. Paul Friedrichs, M.D., inaugural director of the White House Office of Pandemic Preparedness and Response Policy; the Honorable Jill Hruby, former administrator of the National Nuclear Security Administration; and Eric Horvitz, chief scientific officer at Microsoft, who advises on both biodefense and AI.

“Biological threats are a defense problem, not just a public health one. Even with recent progress in detection and response, in particular for novel pathogens, AI means we need to advance the state of the art in biodefense even faster,” said biodefense advisor Hon. Andrew Weber. “It is imperative that we use the tools that are at the frontier of biological design and actively use them to defend against biological threats. What Radical Numerics is building will leap us forward in that mission.”

To see how Omnii designed the patient’s phage, read the Radical Numerics blog post: https://www.radicalnumerics.ai/blog/omnii-1.

To join the team, visit jobs.ashbyhq.com/radical-numerics.

About Radical Numerics

Radical Numerics is an AI research lab building general biological intelligence: AI that can understand, predict and design across the molecular systems of life. Founded by members of the team behind Evo and Evo 2, the company develops frontier biological AI models and holds a dual mandate: advance biological intelligence and strengthen the biosecurity needed to protect it. To learn more, visit radicalnumerics.ai.

Media gallery

About The Author
Business Wire