←
Armin W. Thomas719 posts
Armin W. Thomas avatar

Armin W. Thomas

@athmsx

Co-founder and CTO @RadicalNumerics He/him

◷ Joined August 2016▧ 37 Photos and videos
Eric Nguyen avatar
⟳ Armin W. Thomas retweeted
Eric Nguyen@exnx·Nov 10

I wanted to share some context around the recent discussion of our phage work. A few thoughts on the collaboration and the path forward. Blog: radicalnumerics.ai/blog/phag…

Ben Siranosian avatar
⟳ Armin W. Thomas retweeted
Ben Siranosian@BenSiranosian·Nov 6

It takes a lot more than GPUs to get an AI-designed therapy into a patient. Here, we share some of the process of administering a phage therapy for compassionate use. The partners, manufacturers, and clinicians we work with are key!

Designing phages with AI is one challenge. Getting it to a patient is another. Behind every treatment is a network of scientists, physicians, manufacturers, and partners working together. Our new blog explores the journey from AI-generated DNA to patient care, and what it will take to scale it. Blog: radicalnumerics.ai/blog/phag…

Eric Nguyen avatar
⟳ Armin W. Thomas retweeted
Eric Nguyen@exnx·Nov 6

Designing phages with AI is one challenge. Getting it to a patient is another. Behind every treatment is a network of scientists, physicians, manufacturers, and partners working together. Our new blog explores the journey from AI-generated DNA to patient care, and what it will take to scale it. Blog: radicalnumerics.ai/blog/phag…

Armin W. Thomas avatar
Armin W. Thomas@athmsx·Nov 2

It's been a wild few years, from HyenaDNA to Evo and now Radical Numerics. I'm proud of the science we've done, but even more excited that it's starting to make a real difference in biology. We've barely begun exploring what these models can do. A bit of the backstory here.

A few years ago, we were just a group of researchers chasing unconventional bets in AI. That journey took us from new architectures like Hyena, to Evo and creating the field of generative genomics, to a much bigger realization. We need AI that can understand biology as a whole. We started @RadicalNumerics to pursue that vision. Here’s a conversation with my cofounders on how we got here, why we took the leap, and what we believe comes next.

17:26
Armin W. Thomas avatar
Armin W. Thomas@athmsx·Oct 28

Detecting cancer early from a blood test is hard. We adapted Omnii-1 to data it had never seen in training. It meaningfully improved cancer detection, especially in the hardest cases. Very proud of the team behind this!

1/ With Omnii Diagnostics, we asked whether a multimodal genome language model could help find cancers whose signals in blood are too faint for current classifiers. We adapted Omnii-1 to a biomarker never seen in pretraining. In a test cohort, Omnii's features improved early-stage cancer detection

0:13
Radical Numerics avatar
⟳ Armin W. Thomas retweeted
Radical Numerics@RadicalNumerics·Oct 22

1/10 Introducing Omnii Genome, one of eight Advances built on Omnii-1. A single generalist model now outperforms leading specialized models (AlphaGenome) in linking genetic changes to gene activity and mapping regulatory interactions. These capabilities help make sense of the 98% of the human genome that doesn’t code for proteins How Omnii-1 works:

1:31
Armin W. Thomas avatar
Armin W. Thomas@athmsx·Oct 22

We’re only beginning to see what generative models trained across biological modalities can do. Omnii Genome is a great example: built on Omnii-1, it demonstrates strong performance across very different genomic tasks, while supporting capabilities beyond those of leading specialized models. Really proud of the team that made this happen. Lots more to come!

1/10 Introducing Omnii Genome, one of eight Advances built on Omnii-1. A single generalist model now outperforms leading specialized models (AlphaGenome) in linking genetic changes to gene activity and mapping regulatory interactions. These capabilities help make sense of the 98% of the human genome that doesn’t code for proteins How Omnii-1 works:

1:31
Peter Zhou avatar
⟳ Armin W. Thomas retweeted
Peter Zhou@xtzhou·Oct 20

When we backed @RadicalNumerics, the bet was that one general model could learn the language of biology, the same way LLMs did for text. Today Omnii-1 beats AlphaGenome, designs antibodies and cancer vaccine candidates, and designed a phage covered by @carlzimmer in @NYTimes. All from one model. 🧬

Introducing Omnii-1, our most powerful genome language model, and a major step toward general biological intelligence In a world first, Omnii-1 designed the first AI-generated bacteriophages to be administered to a human patient with an antibiotic-resistant infection under compassionate use

2:35
Eric Nguyen avatar
⟳ Armin W. Thomas retweeted
Eric Nguyen@exnx·Oct 20

Introducing Omnii-1, our most powerful genome language model, and a major step toward general biological intelligence In a world first, Omnii-1 designed the first AI-generated bacteriophages to be administered to a human patient with an antibiotic-resistant infection under compassionate use

2:35
Armin W. Thomas avatar
Armin W. Thomas@athmsx·Oct 20

Omnii-1 is out. We've learned a lot about what happens when you train a single model across biology rather than solving each problem separately. Some of the more surprising results haven't made it into a blog post yet. Incredibly proud of the team that built this!

Introducing Omnii-1, our most powerful genome language model, and a major step toward general biological intelligence In a world first, Omnii-1 designed the first AI-generated bacteriophages to be administered to a human patient with an antibiotic-resistant infection under compassionate use

2:35
Michael Poli avatar
⟳ Armin W. Thomas retweeted
Michael Poli@MichaelPoli6·Oct 20

Today, we are launching Omnii-1, our next-generation multimodal genome language model. Omnii-1 draws on a long line of research breakthroughs in genome language modeling, spanning model architectures, multimodal pretraining, as well as new mid-training & post-training methods tailored to biological data. The launch includes a staggering amount of technical material: 8 full-length technical blogs (and we are just getting started). Proud of the team for this milestone!

Introducing Omnii-1, our most powerful genome language model, and a major step toward general biological intelligence In a world first, Omnii-1 designed the first AI-generated bacteriophages to be administered to a human patient with an antibiotic-resistant infection under compassionate use

2:35
Armin W. Thomas avatar
Armin W. Thomas@athmsx·Oct 16

Excited to welcome @tri_dao as a scientific advisor @RadicalNumerics ! We go way back to our @HazyResearch days, working together on some of the first hybrid models (eg H3 / Hungry Hungry Hippos). Great to have you aboard, Tri!

I'm excited to share that @tri_dao has joined @RadicalNumerics as a scientific advisor! Tri and I were labmates at Stanford's Hazy Research, where he helped pioneer advances in efficient ML systems (FlashAttention). Scaling AI across multimodal biological data demands entirely new approaches to ML systems, and Tri will help guide our research at this frontier. Welcome aboard, Tri! Read more: radicalnumerics.ai/blog/tri-…

Radical Numerics avatar
⟳ Armin W. Thomas retweeted
Radical Numerics@RadicalNumerics·Sep 8

1/5 We pretrained Omnii, a multimodal language model on DNA. Then, we asked it to design a cancer vaccine In new research, we show how Omnii learned to turn a tumor DNA sequence into a personalized mRNA vaccine design. Read more here: radicalnumerics.ai/blog/omni…

Post attachment
Eric Nguyen avatar
⟳ Armin W. Thomas retweeted
Eric Nguyen@exnx·Aug 4

I sat down with @mariogabriele on the Generalist podcast to talk about why DNA is the most complicated and important language ever, how we trained the largest genome language model to master it, and yet, we’ve only scratched the surface of what it can do and understand.

What happens when AI can write the genetic code for life itself? Eric Nguyen (@exnx) is the co-founder of @RadicalNumerics, the lab behind the models that generated the first genome ever created entirely by AI, a bacteriophage that had never existed in nature until it wrote one. He believes DNA should be treated the way language is treated: something a model can read, then write. With more DNA sitting in public databases than there is text on the entire internet, he sees an almost unmined field, one where the models built to make sense of it remain, by his own admission, no more sophisticated than linear regression. Thank you to the partners who make this possible @ahrefs Brand Radar: Find your brand in AI results: ahrefs.com/generalist @brexHQ: The intelligent finance platform: brex.com/mario @Guru_HQ: The AI source of truth for work: (3:35) An overview of Radical Numerics (11:08) Why they started with DNA (26:24) The limits of language models in biology (31:15) Eric’s free-range upbringing and path to his PhD program (41:20) Applying long-context models to DNA and meeting his co-founders (55:30) How better general LLMs benefit Radical Numerics (1:02:05) Making biology more concrete (1:18:09) Final meditations

1:22:14
Mario Gabriele avatar
⟳ Armin W. Thomas retweeted
Mario Gabriele@mariogabriele·Aug 4

What happens when AI can write the genetic code for life itself? Eric Nguyen (@exnx) is the co-founder of @RadicalNumerics, the lab behind the models that generated the first genome ever created entirely by AI, a bacteriophage that had never existed in nature until it wrote one. He believes DNA should be treated the way language is treated: something a model can read, then write. With more DNA sitting in public databases than there is text on the entire internet, he sees an almost unmined field, one where the models built to make sense of it remain, by his own admission, no more sophisticated than linear regression. Thank you to the partners who make this possible @ahrefs Brand Radar: Find your brand in AI results: ahrefs.com/generalist @brexHQ: The intelligent finance platform: brex.com/mario @Guru_HQ: The AI source of truth for work: (3:35) An overview of Radical Numerics (11:08) Why they started with DNA (26:24) The limits of language models in biology (31:15) Eric’s free-range upbringing and path to his PhD program (41:20) Applying long-context models to DNA and meeting his co-founders (55:30) How better general LLMs benefit Radical Numerics (1:02:05) Making biology more concrete (1:18:09) Final meditations

1:22:14
Armin W. Thomas avatar
Armin W. Thomas@athmsx·Jun 23

Super excited about @EngramLab - Stellar team & awesome mission!

Engram @EngramLab Jun 23 Article Introducing Engram: Scaling compute on your context We’re Engram. We’re building AI that learns from you and deeply understands your work. Today’s AI models don’t understand what you do. Not really. Everything models know comes from their training –